System

Abstract

A system is provided.SOLUTION: A system comprising: means for automatically collecting expense data from a data source; means for verifying consistency and completeness of the collected expense data and supplementing as needed; means for analyzing the collected and supplemented data and identifying outliers; means for generating cost reduction suggestions based on the identified outliers; and means for developing a plan for executing the generated suggestions and monitoring their progress and effect.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] The complex cost management challenges facing companies are increasing due to the diversification and globalization of their business operations. Traditional manual data collection and cost analysis methods are inefficient and time-consuming, making it difficult to make quick decisions. Furthermore, there is a need to quickly and accurately identify wasteful costs and abnormal expenses and take appropriate countermeasures, but achieving this requires advanced technology and dedicated systems. To solve these challenges, a system is needed that utilizes advanced AI technology to efficiently collect and analyze data and make improvement proposals in real time. [Means for solving the problem]

[0005] This invention provides a cost optimization system that utilizes AI technology. Specifically, the system includes a means for automatically collecting expense data from data sources, a means for verifying the consistency and completeness of the collected expense data and supplementing it as necessary, a means for analyzing the collected and supplemented data and identifying outliers, a means for generating cost-reduction proposals based on the identified outliers, and a means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness. This system enables efficient corporate cost management, reduces wasteful costs, and achieves optimal operations. Furthermore, by further including a means for categorizing expense data and performing detailed analysis based on specific categories, and a means for prioritizing reduction proposals and formulating an implementation plan based on the priorities, more effective cost optimization can be achieved.

[0006] "Data Source" means an external information source that the cost optimization system connects to to collect expense data.

[0007] "Expense data" refers to data that specifically shows information regarding various costs required for a company's operations.

[0008] "Integrity" refers to data being consistent, consistent, and trustworthy.

[0009] "Complete" refers to the state in which data is free of gaps or omissions and contains all necessary information.

[0010] "Imputation" refers to the process of correcting missing or erroneous data to produce a complete dataset suitable for analysis.

[0011] "Analysis" refers to the process of statistically processing collected data to extract meaningful information.

[0012] An "outlier" is a value that deviates significantly from other data points in a data set.

[0013] A "cost reduction proposal" refers to a proposal that shows specific action plans and measures to reduce waste and improve efficiency.

[0014] "Priority" refers to setting the order of execution for multiple proposals or tasks according to their importance and urgency.

[0015] "Implementation plan" refers to a plan that outlines detailed steps and schedules for specifically implementing proposed cost-cutting measures.

[0016] "Monitoring" refers to the process of continually observing an ongoing process or activity and measuring its progress and effectiveness.

[0017] "AI technology" refers to various technologies and algorithms that use artificial intelligence and are used for data analysis and predictive model generation. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0020] First, the terms used in the following description will be explained.

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0022] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0023] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0039] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[0040] The system of the present invention achieves cost optimization mainly through the following steps.

[0041] 1. Data Collection

[0042] The server automatically collects the company's expense data from data sources such as ERP systems, accounting software, etc. It sends API requests to retrieve the required data.

[0043] Example: A server accesses a company's ERP system and downloads last month's expense report.

[0044] 2. Data cleansing

[0045] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes formats, and corrects invalid values.

[0046] Example: Standardizing the date format of data collected by the server and correcting incorrect values.

[0047] 3. Cost Analysis

[0048] The server performs basic statistical processing on the organized data to identify outliers, and also performs trend analysis on past data to predict future costs.

[0049] Example: A server analyzes energy consumption data from the past 12 months and predicts consumption trends for the next year.

[0050] 4. Categorization

[0051] The server automatically categorizes expense data and allows users to analyze it by specific categories, with the option to manually adjust the categorization as needed.

[0052] Example: The server categorizes collected expense data by department and project, and the user adjusts it appropriately.

[0053] 5. Identify waste

[0054] The server quickly identifies wasteful costs and excessive spending based on outliers, which are then evaluated by AI algorithms to highlight areas for improvement.

[0055] Example: A server identifies unnecessarily high communication costs and analyzes the cause.

[0056] 6. Cost-cutting measures

[0057] The server generates specific cost-cutting proposals based on the results of waste identification, and also simulates the effects of the proposals in advance to set priorities.

[0058] Example: The server proposes specific measures for optimizing energy usage and prioritizes them based on their importance.

[0059] 7. Implementation Plan

[0060] The server then creates a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments.

[0061] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[0062] 8. Cost reduction implementation

[0063] The terminal carries out the execution activities based on the formulated plan, and the server monitors this in real time and records the progress.

[0064] Example: A user implements new energy usage guidelines and a server monitors their progress.

[0065] 9. Confirmation of effectiveness

[0066] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0067] Example: A server reviews the effectiveness of new energy usage methods and suggests further improvements.

[0068] Each step of this system automates a company's cost management and improves operational efficiency by accurately and quickly identifying and eliminating waste.

[0069] The processing flow will be explained below.

[0070] Step 1:

[0071] The server accesses a data source, such as an ERP system or accounting software, and sends an API request to retrieve expense data, which includes verifying credentials and specifying the time period to retrieve.

[0072] Example: A server accesses a company's ERP system and downloads last month's expense data using an API.

[0073] Step 2:

[0074] The server checks the integrity and completeness of the retrieved data, adapting to the data format and content and identifying missing or outlier values.

[0075] Example: Standardizing the date format and number of digits in data collected by the server and detecting inappropriate values.

[0076] Step 3:

[0077] The server uses statistical models and predictive algorithms to impute missing values, and also standardizes and cleans the data.

[0078] Example: A server uses a predictive model to fill in missing energy usage data and correct improper formatting.

[0079] Step 4:

[0080] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[0081] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[0082] Step 5:

[0083] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[0084] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[0085] Step 6:

[0086] The server automatically categorizes expense data by category, allowing users to perform detailed analysis by specific category and sorting the data.

[0087] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[0088] Step 7:

[0089] The server identifies wasteful costs using an outlier detection algorithm, quickly uncovering expenditure items that fall outside of normal ranges.

[0090] Example: A server detects extremely high communication costs and analyzes the background.

[0091] Step 8:

[0092] The server generates specific cost-cutting proposals based on the identified waste, and the AI ​​automatically devise effective improvement methods and provides them to the user.

[0093] Example: Propose specific measures to improve the energy efficiency of servers.

[0094] Step 9:

[0095] The server assigns a priority to each suggestion, allowing the user to execute the most effective suggestion first.

[0096] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[0097] Step 10:

[0098] The server creates an action plan based on the reduction proposals, generates a plan including detailed procedures and schedules, and presents it to the user.

[0099] Example: The server creates a specific action plan for reducing energy usage, which the user confirms.

[0100] Step 11:

[0101] The terminal executes cost-cutting activities based on the established plan, and the server monitors this process in real time and records the progress.

[0102] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[0103] Step 12:

[0104] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0105] Example: A server reviews the effectiveness of new energy usage methods and suggests further improvements.

[0106] As described above, by performing specific operations at each step, a system is provided that realizes cost optimization for a company.

[0107] Example 1

[0108] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0109] Current corporate cost management systems often involve manual data collection and analysis processes, resulting in a lack of efficiency. It also makes it difficult to detect wasteful costs and outliers early on, making it difficult to propose effective cost-cutting measures. Furthermore, the consistency and completeness of the collected data cannot be guaranteed, making it difficult to obtain reliable analysis results. To solve these problems and improve corporate operational efficiency, automated data collection, cleansing, and analysis processes are needed.

[0110] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0111] In this invention, the server includes means for automatically collecting expense data from data sources, means for verifying the consistency and completeness of the collected expense data and imputing missing values ​​using a predictive model, means for performing statistical processing on the collected and imputed data to identify outliers, means for generating cost reduction proposals using an AI algorithm based on the identified outliers, means for simulating the effects of the generated proposals in advance and setting priorities, and means for formulating action plans based on the priorities and monitoring their progress and effectiveness in real time, thereby enabling companies to efficiently collect and analyze data and quickly identify and reduce wasteful costs.

[0112] "Data source" refers to the systems and software that provide a company's expense data.

[0113] "Expense data" refers to data that includes information about expenses incurred by a company in various activities.

[0114] "Integrity" refers to a state in which data is consistent and free of contradictions.

[0115] "Complete" refers to the state in which data contains all necessary information.

[0116] A "predictive model" refers to an algorithm or mathematical model that estimates future values ​​based on trends and patterns in data.

[0117] "Statistical processing" refers to methods for summarizing and characterizing data as part of data analysis.

[0118] An "outlier" is a value in the data that is significantly different from the other values.

[0119] "AI algorithm" refers to a data processing method based on machine learning and artificial intelligence.

[0120] "Cost reduction proposals" refer to specific methods and measures for reducing unnecessary costs.

[0121] "Simulation" refers to the process of testing the effectiveness of a proposal in a virtual environment.

[0122] "Priority" refers to ranking multiple items or proposals according to their importance.

[0123] "Implementation plan" refers to a plan for specifically implementing the proposed cost reduction measures.

[0124] "Real-time" refers to a state in which processing and data collection occur immediately.

[0125] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[0126] The main hardware of the system is a server, which automatically collects expense data from data sources such as a company's ERP system and accounting software. Specifically, it retrieves the necessary data from these sources by sending API requests. The server also checks the consistency and completeness of the collected data and imputes missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The server also standardizes the data format and corrects inappropriate values.

[0127] The server then performs statistical operations on the organized data to identify outliers. For example, it uses Pandas to calculate the standard deviation of the data frame to identify outliers. The server also performs trend analysis on the historical data and uses forecasting models such as ARIMA models to forecast future costs.

[0128] The server then automatically categorizes expense data by department and project, allowing users to manually adjust this as needed. It uses Python regular expressions to parse expense record descriptions and classify them into the appropriate categories. Users can review and modify the categorization results using a web interface.

[0129] To identify wasteful costs, the server uses AI algorithms (e.g., random forests) to assess wasteful costs and excessive spending based on outliers, allowing for quick identification of areas for improvement.

[0130] The server also generates specific cost-reduction proposals based on the results of identifying waste. The generated proposals have the ability to simulate their effects in advance and set priorities. For example, the server generates a proposal for installing new lighting equipment and simulates its energy-saving effects. It also calculates the expected cost savings for each proposal and creates a priority list.

[0131] Furthermore, the server creates an action plan based on the priority, which the user can review and ultimately adjust.Specific action plans for each department are created in Excel files, which the user can review and modify via a web interface.

[0132] Finally, the device carries out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. For example, a user implements new energy usage guidelines, and the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard. After this implementation, the server re-analyzes the post-improvement data to confirm the effectiveness of the implementation of the new energy usage guidelines.

[0133] Example prompts to input to a generative AI model:

[0134] "Please explain a program that collects expense data from an ERP system, cleanses the data by imputing missing values ​​with a predictive model, identifies outliers through statistical processing and trend analysis, generates optimization proposals, and formulates specific implementation plans."

[0135] Through this system, companies can streamline cost management and reduce wasteful spending.

[0136] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0137] Step 1:

[0138] Data collection

[0139] The server automatically collects expense data from ERP systems and accounting software. Specifically, the server sends API requests to retrieve the required data from these data sources. The input is the API endpoint of each data source, and the output is the collected expense data. For example, the server sends a request such as "GET / api / expenses?month=2023-09" to the ERP system, receives expense data as a response, and stores it in the database.

[0140] Step 2:

[0141] Data Cleansing

[0142] The server checks the consistency and completeness of the collected expense data and imputes missing values. Specifically, it predicts missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The input is the collected raw expense data, and the output is cleansed data with missing values ​​imputed and a unified format. The server runs a script to unify the date format to "YYYY-MM-DD" and corrects inappropriate values.

[0143] Step 3:

[0144] Cost Analysis

[0145] The server performs statistical processing on the cleansed data to identify outliers. The input is the cleansed expense data, and the output is an analysis showing outliers and trends. For example, the server uses Pandas to calculate the standard deviation of the data frame to identify outliers. It also uses an ARIMA model to forecast future cost trends using data from the past 12 months.

[0146] Step 4:

[0147] Category Classification

[0148] The server categorizes expense data by department or project, allowing users to manually adjust as needed. The input is the analyzed expense data, and the output is the categorized expense data. The server uses Python regular expressions to parse the expense record description and classify it into the appropriate category. Users can view the categorization results through a web interface and make manual adjustments.

[0149] Step 5:

[0150] Identifying waste

[0151] The server identifies wasteful costs and excessive expenditures based on outliers. The input is the analysis results, including outliers, and the output is a list of wasteful costs. Specifically, the server uses an AI algorithm (e.g., random forest) to evaluate wasteful expenditures. The server detects abnormally high communication costs and displays them in a histogram.

[0152] Step 6:

[0153] cost-cutting measures

[0154] The server generates cost-saving proposals based on the results of waste identification. The input is a list of wasteful costs, and the output is specific cost-saving proposals. The server generates proposals for installing new lighting equipment and simulates their energy-saving effects. It calculates the expected cost savings for each proposal and creates a priority list.

[0155] Step 7:

[0156] Implementation Plan

[0157] The server creates an action plan for cost-cutting proposals, which the user can review and adjust. The input is the cost-cutting proposals, and the output is a specific implementation plan. For example, the server creates a specific action plan for each department in an Excel file, and the user can review the plan via a web interface and modify it as needed.

[0158] Step 8:

[0159] Cost reduction implementation

[0160] The terminals carry out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. The input is the implementation plan, and the output is progress data of the implementation activities. For example, when a user puts new energy usage guidelines into practice, the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard.

[0161] Step 9:

[0162] Confirmation of effectiveness

[0163] The server re-analyzes the post-implementation data and confirms the effects of the improvements. The input is the post-implementation data, and the output is the analysis of the effects. The server analyzes the data after the new energy usage guidelines are implemented and displays the effects in a graph. The server then presents the user with a report showing the potential for further cost reductions.

[0164] Through this series of steps, companies can streamline cost management and quickly identify and reduce wasteful spending.

[0165] (Application example 1)

[0166] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0167] Corporate cost management is complex, time-consuming, and labor-intensive, often resulting in a lot of unnecessary expenses. Furthermore, logistics centers often lack real-time information on resource usage, making efficient cost reduction difficult. The present invention aims to solve these problems by providing a system for efficiently managing corporate expenses and reducing unnecessary costs.

[0168] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0169] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost reduction proposals based on the identified outliers, means for formulating plans for implementing the generated proposals and monitoring their progress and effectiveness, and means for understanding the resource usage status of the logistics center in real time via a smartphone application and quickly identifying and reducing unnecessary costs. This automates corporate expense management and enables real-time cost management and reduction at logistics centers.

[0170] A "data source" is an information system or database that provides a company's expense data.

[0171] "Expense data" refers to data that includes all expense information related to the day-to-day running of a company.

[0172] "Integrity" refers to a state in which data is consistent and free of contradictions.

[0173] "Complete" refers to the state in which all data is present without any missing data.

[0174] "Completion" refers to the process of filling in missing data using predictive models, etc.

[0175] An "outlier" is a value that is significantly different from the rest of the data and is outside the range of what is normally expected.

[0176] A "cost reduction proposal" is a specific measure or action plan that should be implemented to reduce wasteful expenses.

[0177] "Plan" refers to the design of specific steps and actions to implement a proposal.

[0178] "Progress" is a state or condition that indicates how far a planned action has been carried out.

[0179] "Effectiveness" is the result that shows how much cost savings the implemented proposals and plans actually achieved.

[0180] A "smartphone application" is a software program that runs on a smartphone and acts as a user interface to provide real-time information and enable operation.

[0181] A "logistics center" is a base for storing, managing, and delivering goods, and is a facility that carries out efficient logistics activities.

[0182] "Resource usage" refers to the utilization of energy, labor, equipment, etc. within a logistics center.

[0183] This invention relates to a system for efficiently managing corporate costs and resource usage at a logistics center. This system can be realized using a smartphone application and a server.

[0184] 1. System Configuration

[0185] The server automatically collects expense data from data sources and checks the consistency and completeness of the collected expense data. It supplements missing data as needed and analyzes the collected data to identify outliers. It generates cost-cutting proposals based on the outliers, develops detailed plans for implementing the proposals, and monitors their progress and effectiveness. Furthermore, a smartphone application allows users to grasp the distribution center's resource usage in real time, quickly identifying and reducing unnecessary costs.

[0186] 2. Hardware and Software

[0187] The server uses the following hardware and software:

[0188] Hardware: A server with a powerful processor and ample storage capacity

[0189] Software: ERP system and accounting software APIs for data collection, pandas and NumPy for data cleansing, TensorFlow and Scikit-learn for statistical processing and outlier detection, Flask / Django as web application frameworks, Firebase / Firestore as real-time databases

[0190] The smartphone application has the following features:

[0191] Data collection: Collect data from ERP systems, sensors in distribution centers, and barcode readers.

[0192] Real-time monitoring: View resource usage in your distribution center in real time

[0193] Notification function: Alerts when unnecessary costs or abnormal values ​​occur

[0194] 3. Program Processing Overview

[0195] The server cleanses expense data collected from data sources for consistency and completeness, and uses predictive models to fill in any missing data. It then performs basic statistical processing on the cleansed data to identify outliers. Based on the identified outliers, an AI algorithm generates cost-saving proposals and performs simulations to verify their effectiveness. Finally, it develops prioritized cost-saving plans, and monitors their progress and effectiveness in real time.

[0196] 4. Examples and prompts

[0197] For example, if you want to predict consumption trends for the next year based on energy consumption data from a logistics center over the past 12 months, you can input the following prompt into the generative AI model:

[0198] "Given the last 12 months of energy consumption data for a distribution center, please generate a model that will predict consumption trends for the next fiscal year. Also, please generate and prioritize specific proposals for cost reduction."

[0199] Using this prompt, the AI ​​model generates output through the following steps:

[0200] 1. Preprocessing of input data

[0201] 2. Trend Forecasting

[0202] 3. Outlier detection

[0203] 4. Cost reduction proposals

[0204] 5. Prioritization

[0205] This system will significantly improve the efficiency of cost management at logistics centers and enable real-time cost reductions.

[0206] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0207] Step 1:

[0208] Data collection

[0209] The server collects expense and resource usage data from the ERP system, sensors in the distribution center, and barcode readers. As input, it sends API requests to retrieve the required data. For example, it accesses the ERP system to download expense reports for the past 12 months and retrieves real-time energy consumption data from sensors. As output, these data are passed to subsequent processing steps.

[0210] Step 2:

[0211] Data Cleansing

[0212] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes the date format of the data, and corrects improper values. It receives the collected raw data as input and produces cleansed and organized data as output.

[0213] Step 3:

[0214] Cost Analysis

[0215] The server performs basic statistical processing on the cleaned data to identify outliers. It also performs trend analysis on past data to predict future costs. It takes the cleansed data as input and provides outliers and predictions of future cost trends as output. For example, it analyzes energy consumption data from the past 12 months to predict consumption trends for the next year.

[0216] Step 4:

[0217] Category Classification

[0218] The server automatically categorizes expense data and allows users to analyze it by specific categories. Users can also manually adjust the categorization results as needed. It uses the analyzed data as input and generates categorized data as output. For example, categorizing expense data by department or project.

[0219] Step 5:

[0220] Identifying waste

[0221] The server quickly identifies wasteful costs and excessive expenditures based on outliers. AI algorithms evaluate these and identify areas for improvement. It receives outlier data as input and provides the results of wasteful cost identification as output. For example, it identifies unnecessarily high communication costs and analyzes the causes.

[0222] Step 6:

[0223] cost-cutting measures

[0224] The server generates specific cost-saving proposals based on the waste identification results. It also simulates the effectiveness of the proposals in advance and sets priorities. It uses the waste identification results and related data as input and provides cost-saving proposals and their predicted effects as output. For example, it proposes specific measures for optimizing energy use and prioritizes the proposals based on their importance.

[0225] Step 7:

[0226] Implementation Plan

[0227] The server then develops a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments. It uses the cost-saving proposals as input and provides a detailed implementation plan as output. For example, an energy reduction plan can be developed, which the user can review and adjust to fit their office hours.

[0228] Step 8:

[0229] Cost reduction implementation

[0230] The device performs the execution activities based on the formulated plan. The server monitors this in real time and records the progress. It uses the implementation plan as input and generates execution progress data as output. For example, a user implements new energy usage guidelines and the server monitors their progress.

[0231] Step 9:

[0232] Confirmation of effectiveness

[0233] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements. If necessary, it will propose further optimizations. It uses the execution result data as input and provides the results of the effectiveness check and further improvement suggestions as output. For example, it can confirm the effectiveness of a new energy usage method and propose further improvements.

[0234] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0235] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management. In particular, it is characterized by combining an emotion engine that recognizes the user's emotions and providing customized suggestions according to the user's emotions.

[0236] The system of the present invention includes the following elements:

[0237] 1. Data Collection

[0238] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data.

[0239] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[0240] 2. Data cleansing

[0241] The server checks the consistency and completeness of the retrieved data, imputes missing values ​​using predictive algorithms, and standardizes the format.

[0242] Example: The server corrects mismatched date formats and missing values.

[0243] 3. Cost Analysis

[0244] The server performs statistical processing to identify outliers in expenses and then performs trend analysis to predict future costs.

[0245] Example: A server analyzes energy consumption data and predicts consumption trends for the next year.

[0246] 4. Categorization

[0247] The server automatically categorizes expense data and allows users to review and adjust the details.

[0248] Example: A server categorizes expenses by department or project.

[0249] 5. Identify waste

[0250] The server uses an anomaly detection algorithm to quickly identify wasteful costs, and AI evaluates the potential for improvement.

[0251] Example: A server identifies high communication charges and analyzes their causes.

[0252] 6. Cost reduction proposals

[0253] The server generates specific cost-cutting proposals based on the identified waste, simulates the effectiveness of the proposals, and sets priorities.

[0254] Example: Propose specific measures to improve the energy efficiency of a server and list them in order of importance.

[0255] Additionally, the system of the present invention incorporating the emotion engine includes the following elements:

[0256] 1. Emotion recognition

[0257] The server uses an emotion engine to recognize the user's emotions and analyzes the user's writings and comments to identify the user's emotional state.

[0258] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[0259] 2. Emotion-based personalized recommendations

[0260] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. For example, if the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[0261] Example: When a user is feeling stressed, the server prioritizes suggestions for reducing stress that are relatively easy to implement.

[0262] 3. Emotional Feedback

[0263] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[0264] Example: The server displays a dashboard showing the user's current emotional state and provides suggestions for improvement based on that.

[0265] For example, if a company's human resources manager is feeling stressed, the system will recognize that emotion and first offer relatively simple and effective improvement suggestions, allowing the manager to start with suggestions that are easy to implement. The system also monitors the effectiveness of the suggestions after implementation and, if further suggestions are needed, suggests next steps based on the manager's emotional state.

[0266] In this way, the present invention aims to make corporate cost management more effective and user-friendly by combining AI technology and emotion recognition technology.

[0267] The processing flow will be explained below.

[0268] Step 1:

[0269] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data, which includes verifying credentials and specifying the time period to retrieve.

[0270] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[0271] Step 2:

[0272] The server checks the integrity and completeness of the retrieved data, adapts to the data format and content, identifies missing or outlier values, and imputes data as needed.

[0273] Example: Standardizing the date format and number of digits in data collected by the server, and detecting and correcting inappropriate values.

[0274] Step 3:

[0275] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[0276] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[0277] Step 4:

[0278] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[0279] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[0280] Step 5:

[0281] The server automatically categorizes expense data by category, allows users to perform detailed analysis for each specific category, and allows users to manually adjust the categorization results as needed.

[0282] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[0283] Step 6:

[0284] The server uses AI technology to identify outliers, using predictive algorithms to spot spending items that fall outside of the normal range.

[0285] Example: A server detects extremely high communication costs and analyzes the background.

[0286] Step 7:

[0287] The server generates specific cost-saving proposals based on the identified outliers, simulates the effects of the proposals, and provides the proposals in an easy-to-implement format for the user.

[0288] Example: Propose specific measures to improve the energy efficiency of servers.

[0289] Step 8:

[0290] The server prioritizes the generated reduction proposals, allowing the user to implement the most effective proposals first.

[0291] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[0292] Step 9:

[0293] The server creates a detailed implementation plan based on the reduction proposals, including specific steps and schedules, which the user can review and adjust.

[0294] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[0295] Step 10:

[0296] The terminal executes cost-cutting activities based on the established plan, and the server monitors the progress in real time and provides feedback to the user.

[0297] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[0298] Step 11:

[0299] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0300] Example: A server reviews the effectiveness of new energy usage methods and suggests further improvements.

[0301] (A system incorporating an emotion engine)

[0302] Step 12:

[0303] The server uses an emotion engine to recognize the user's emotions by analyzing the user's posts and comments to identify their emotional state.

[0304] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[0305] Step 13:

[0306] The server customizes the content and priorities of cost-saving proposals based on the user's recognized emotions, giving priority to proposals that are less burdensome, taking into account the user's emotional state.

[0307] Example: When a server detects a user is feeling stressed, it prioritizes suggestions for reducing stress that are relatively easy to implement.

[0308] Step 14:

[0309] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[0310] Example: The server displays a dashboard showing the user's current emotional state and provides appropriate improvement suggestions based on that.

[0311] In this way, a system that combines an emotion engine can make flexible cost optimization proposals that take into account the user's emotional state, aiming to make corporate cost management more efficient and user-friendly.

[0312] Example 2

[0313] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0314] Conventional cost management systems collect and analyze expense data, generate cost-reduction proposals, and formulate implementation plans, but they are unable to provide effective proposals that take into account the user's emotional state. This can lead to stress and difficulty in implementing the proposals. Therefore, there is a need for a system that recognizes the user's emotional state and provides customized proposals based on that state.

[0315] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0316] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost-cutting proposals based on the identified outliers, means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness, means for identifying the user's emotional state using an emotion engine that recognizes the user's emotions, means for customizing the content and priorities of the cost-cutting proposals based on the user's emotional state, and means for visually providing feedback on the user's emotional state, thereby enabling effective and less burdensome cost-cutting proposals to be made in accordance with the user's emotional state.

[0317] "Data sources" refers to the systems and applications that store a company's expense data, including ERP systems and accounting software.

[0318] "Expense data" is data that contains detailed information about the costs a company incurs on a day-to-day basis.

[0319] "Integrity" means that data is accurate, consistent, and does not contradict each other.

[0320] "Complete" means that the data contains all necessary information and is free of missing parts.

[0321] "Imputation" refers to filling in missing information in data using prediction or imputation algorithms.

[0322] An "outlier" is a value that is extremely outlier compared to other values ​​in a data set, and often includes errors or exceptions.

[0323] A "cost reduction proposal" refers to a proposal that outlines specific action items and strategies for reducing a company's expenses.

[0324] "Plan" refers to a document that contains specific steps and timelines for implementing the generated cost reduction proposals.

[0325] "Progress" refers to the status of the implementation of planned specific actions, and indicates how much progress has been made over time.

[0326] "Effects" refer to the results obtained as a result of implementing a plan, and specifically include the degree of cost reduction and improved efficiency.

[0327] "Emotion engine" refers to software equipped with artificial intelligence technology to recognize and analyze a user's emotional state.

[0328] "Emotional state" refers to the psychological state such as stress or satisfaction felt by the user.

[0329] "Feedback" refers to providing the analysis results and suggestions to the user visually or in other ways, allowing the user to check their own status and the suggestions.

[0330] This invention is a system that utilizes AI technology to improve the efficiency of corporate cost management, and is characterized by its incorporation of an emotion engine that recognizes user emotions. To implement this system, hardware and software including the following elements are used.

[0331] 1. Data Collection

[0332] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data. Specifically, the server sends an HTTP request and receives data in JSON format from the ERP system or accounting software.

[0333] Example: A server sends a request to "https: / / api.erp.example.com / v1 / expenses?month=last" to retrieve last month's expense data from a company's ERP system.

[0334] 2. Data cleansing

[0335] The server checks the integrity and completeness of the data retrieved, imputes missing values ​​using predictive algorithms, and standardizes formats, including detecting duplicate entries in the database and using a reverse dictionary to infer missing dates.

[0336] Example: A server corrects inconsistencies in date formats between expense data entries and imputes missing date fields with a predictive algorithm.

[0337] 3. Cost Analysis

[0338] The server performs statistical processing to identify outliers in expenses and forecast future costs. Specifically, it uses the Pandas library to detect outliers and the Prophet model to forecast future cost trends.

[0339] Example: A server analyzes energy consumption data and predicts energy consumption trends for the next year.

[0340] 4. Categorization

[0341] The server automatically categorizes expense data into categories and allows users to review and adjust the details, including using machine learning models to categorize data into specific categories.

[0342] Example: A server organizes expense data into categories such as communication expenses, travel expenses, and advertising expenses.

[0343] 5. Identify waste

[0344] The server uses anomaly detection algorithms to identify wasteful costs and evaluate potential areas for improvement, using techniques such as Isolation Forest and One-Class SVM.

[0345] Example: The server identifies communication expenses that are significantly higher than average as an anomaly and analyzes the details.

[0346] 6. Cost reduction proposals

[0347] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[0348] Example: A server generates a list of proposals for replacing lighting with LEDs to improve energy efficiency, based on estimated results.

[0349] 7. Emotion recognition

[0350] The server uses an emotion engine to recognize user emotions and identify the emotional state from user posts and statements, using a natural language processing model (e.g., BERT).

[0351] Example: A server analyzes a user's message "Work has been very stressful lately" and determines that the user is feeling stressed.

[0352] 8. Emotion-based personalized recommendations

[0353] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. If the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[0354] Example: When a user feels stressed, the server displays a low-impact reduction suggestion such as "reduce the frequency of meetings."

[0355] 9. Emotional Feedback

[0356] The server provides visual feedback of the user's emotional state, allowing the user to check the state and select suggestions using a visualization tool.

[0357] Example: The server displays the user's emotional state on a dashboard as "Stress level: High" and provides suggestions for improvement based on that.

[0358] This system allows companies to achieve efficient and user-friendly cost management, allowing users to receive cost-saving suggestions that take their emotional state into account and are easy to implement.

[0359] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0360] Program processing flow

[0361] Step 1: Data collection

[0362] The server contacts the data source and sends an API request to retrieve expense data.

[0363] Input: API endpoint URL and authentication information configured on the server.

[0364] Specific behavior: The server sends an HTTP request to an endpoint such as "https: / / api.erp.example.com / v1 / expenses?month=last".

[0365] Output: Expense data in JSON format returned from your ERP system or accounting software.

[0366] Step 2: Data cleansing

[0367] The server checks the consistency and completeness of the data it retrieves, fills in missing values, and standardizes the format.

[0368] Input: Captured expense data in JSON format.

[0369] What it does: The server uses a database library to read data, detects duplicate entries, fills in missing data with a predictive algorithm, and standardizes data formats (for example, standardizing date formats to "YYYY-MM-DD").

[0370] Output: A clean dataset that has been checked for consistency and completeness, imputed for missing values, and has a uniform format.

[0371] Step 3: Cost analysis

[0372] The server analyzes the cleansed data to identify outliers and predict future costs.

[0373] Input: The cleansed dataset.

[0374] What it does: The server uses a statistical library to detect outliers in the data and a time series forecasting model (e.g., Prophet) to predict future cost trends.

[0375] Output: Identified outliers and future cost forecast data.

[0376] Step 4: Categorization

[0377] The server automatically categorizes expense data and allows users to review and adjust the details.

[0378] Input: The cleansed dataset.

[0379] What it does: The server uses machine learning models to categorize expense data by category (e.g., communication, travel, advertising).

[0380] Output: Expense data broken down by category.

[0381] Step 5: Identify waste

[0382] The server uses an anomaly detection algorithm to identify wasteful costs and evaluate potential for improvement.

[0383] Input: Expense data broken down by category.

[0384] How it works: The server uses outlier detection algorithms such as Isolation Forest and One-Class SVM to identify unreasonably high expenditures.

[0385] Output: Identified waste costs and a detailed report.

[0386] Step 6: Cost reduction proposal

[0387] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[0388] Input: Detailed data on waste costs.

[0389] Specific operation: The server generates improvement measures using heuristic rules and verifies their effectiveness using a simulation model.

[0390] Output: Prioritized cost-saving proposals and their predicted effectiveness.

[0391] Step 7: Emotion Recognition

[0392] The server uses an emotion engine to recognize the user's emotions and analyzes the user's emotional state from their statements and writings.

[0393] Input: User message and feedback data.

[0394] What happens: The server analyzes the text data using a natural language processing model (e.g., BERT) to identify the user's emotional state.

[0395] Output: User's emotional state report.

[0396] Step 8: Customize your suggestions based on emotions

[0397] The server customizes the suggestions and priorities based on the user's emotions recognized.

[0398] Input: User emotional state report, cost reduction suggestions.

[0399] Specific operation: The server selects and customizes suggestions that are easy to implement and have high priority based on the user's emotional state.

[0400] Output: Customized cost-saving proposals.

[0401] Step 9: Emotional Feedback

[0402] The server visually feeds back the user's emotional state, allowing the user to confirm the state and select suggestions.

[0403] Input: customized cost-saving suggestions, user emotional state report.

[0404] Specific operation: The server uses visualization tools such as a dashboard to display the user's emotional state and suggestions.

[0405] Output: Emotional feedback and improvement suggestions on a visualized dashboard.

[0406] (Application example 2)

[0407] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0408] Conventional cost management systems collect and analyze expense data, identify outliers, and generate cost-reduction proposals. However, these processes do not take into account the user's emotional state, often resulting in stress. Furthermore, in logistics centers, worker comfort and efficiency are important, and workload adjustments based on individual emotional states are required. However, current systems lack an effective means of doing this, resulting in insufficient improvements in work efficiency and reductions in worker stress.

[0409] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost reduction proposals based on the identified outliers, means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness, means for determining the user's emotional state using emotion recognition technology and customizing the content and priority of the proposals based on the determined emotional state, and means having a human interface for presenting the customized proposals to the user. This enables efficient cost management that takes the user's emotional state into consideration, reducing worker stress and improving work efficiency at logistics centers.

[0410] "Data sources" refers to information sources such as ERP systems or accounting software that store a company's expense data.

[0411] "Expense data" is detailed information about the costs a company incurs on a daily basis, including travel, communication, and energy costs.

[0412] "Integrity" refers to data that is consistently accurate and consistent across different data sets.

[0413] "Complete" refers to the state in which all data is available and there are no missing values ​​or incomplete information.

[0414] "Imputation" refers to the process of estimating and adding missing data to complete a dataset.

[0415] An "outlier" is a data value that deviates from normal patterns or acceptable ranges.

[0416] "Cost reduction proposals" refer to specific action plans and measures to reduce a company's expenses.

[0417] A "plan" is a schedule that defines the steps and resources needed to achieve a specific goal.

[0418] "Progress" refers to the status of execution and degree of success of a plan or project.

[0419] "Effect" refers to the results or outcomes that result from a certain action or measure.

[0420] "Emotion recognition technology" refers to technology that identifies a user's emotional state (e.g., stress, joy, anger, etc.) from facial expressions and vocal tone.

[0421] "User" refers to the person who operates the system and manages and analyzes expense data.

[0422] "Emotional state" refers to a user's current psychological state as determined by emotion recognition technology.

[0423] "Human interface" refers to the interface through which a user interacts with a system and exchanges information.

[0424] This invention provides a system aimed at efficient cost management and worker stress reduction in logistics centers. This system utilizes AI technology and emotion recognition technology to make suggestions based on the user's emotional state. Specific embodiments are described below.

[0425] The system consists of a server and smart glasses as its main components. The server automatically collects expense data from data sources such as a company's ERP system or accounting software, including the means to retrieve the required data using API requests.

[0426] The server checks the consistency and completeness of collected expense data, fills in missing values ​​with a predictive algorithm, and standardizes the format, thereby ensuring data quality.

[0427] The server then performs statistical processing and analyses on the collected and stored data, identifying outliers and conducting trend analysis to predict future costs, which in turn generates cost-saving recommendations.

[0428] The server uses emotion recognition technology to collect data through the smart glasses to determine the worker's emotional state. Specifically, it uses facial recognition and voice analysis to determine the worker's emotional state from their facial expressions and tone. Based on the determined emotional state, the server customizes the content and priorities of suggestions.

[0429] The customized suggestions are sent from the server to the smart glasses and presented to the worker through a human interface, so that suggestions for a lighter workload are displayed preferentially to workers who are feeling stressed.

[0430] The server uses TensorFlow and OpenCV to build emotion recognition models, Flask framework to manage data on the server side, and SQLite to operate a local database. Communication with the smart glasses is via Bluetooth Low Energy (BLE).

[0431] For example, if a worker at a logistics center is feeling stressed, the system will recognize that emotion and suggest lighter tasks first, such as avoiding carrying heavy items and prioritizing stocking lighter items, thereby reducing the worker's stress and allowing them to work efficiently.

[0432] Below is an example of a prompt sentence that uses a generative AI model to design suggestions based on the worker's emotions.

[0433] Example prompt sentence:

[0434] Consider a graphical interface design that uses an emotion engine to identify the emotions of workers and suggest lighter tasks to stressed workers. In particular, be sure to consider the use of colors and layout to reduce stress.

[0435] In this way, this system simultaneously improves the cost performance and work efficiency of logistics centers through efficient management of expense data and suggestions based on the emotional state of workers.

[0436] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0437] Step 1:

[0438] The server collects expense data from data sources, specifically, from ERP systems and accounting software using API requests. The input is the access information of the data source based on the API request, and the output is the collected raw expense data.

[0439] Step 2:

[0440] The server checks the collected expense data for consistency and completeness and imputes it if necessary, correcting inconsistent date formats and imputing missing values ​​using predictive algorithms. The input is the raw expense data collected, and the output is cleaned data that has been checked for consistency and completeness.

[0441] Step 3:

[0442] The server analyzes the collected and imputed data to identify outliers. It performs statistical processing to detect outliers that deviate from the normal range. The input is the cleaned expense data, and the output is a list of outliers.

[0443] Step 4:

[0444] The server generates cost reduction proposals based on the identified outliers. It analyzes the causes of the outliers and proposes specific cost reduction measures accordingly. The input is a list of outliers, and the output is the generated cost reduction proposals.

[0445] Step 5:

[0446] The server uses emotion recognition technology to determine the worker's emotional state. It uses the camera and microphone in the smart glasses to perform facial recognition and voice analysis to identify the worker's emotional state. The input is real-time video and audio data, and the output is the determined emotional state.

[0447] Step 6:

[0448] The server customizes the content and priority of the proposals based on the determined emotional state. If the user is feeling stressed, it prioritizes proposals that are less stressful. The input is the determined emotional state and the generated cost-saving proposals, and the output is the customized proposals.

[0449] Step 7:

[0450] The server sends the customized suggestions to the smart glasses for presentation to the worker. The suggestions are displayed to the worker via a human interface using Bluetooth Low Energy (BLE). The input is the customized suggestions, and the output is the suggestions displayed on the smart glasses display.

[0451] Step 8:

[0452] The server develops an implementation plan for the proposals and monitors their progress and effectiveness. It evaluates the results of implementing the proposed cost-saving measures and obtains feedback. The inputs are the implemented proposals and their results data, and the outputs are progress reports and effectiveness evaluation reports.

[0453] By carrying out the above processing steps, efficient cost management and stress reduction for workers at the logistics center can be achieved.

[0454] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0455] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0456] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0457] [Second embodiment]

[0458] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0459] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0460] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0461] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0462] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0463] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0464] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0465] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0466] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0467] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0468] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0469] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0470] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[0471] The system of the present invention achieves cost optimization mainly through the following steps.

[0472] 1. Data Collection

[0473] The server automatically collects the company's expense data from data sources such as ERP systems, accounting software, etc. It sends API requests to retrieve the required data.

[0474] Example: A server accesses a company's ERP system and downloads last month's expense report.

[0475] 2. Data cleansing

[0476] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes formats, and corrects invalid values.

[0477] Example: Standardizing the date format of data collected by the server and correcting incorrect values.

[0478] 3. Cost Analysis

[0479] The server performs basic statistical processing on the organized data to identify outliers, and also performs trend analysis on past data to predict future costs.

[0480] Example: A server analyzes energy consumption data from the past 12 months and predicts consumption trends for the next year.

[0481] 4. Categorization

[0482] The server automatically categorizes expense data and allows users to analyze it by specific categories, with the option to manually adjust the categorization as needed.

[0483] Example: The server categorizes collected expense data by department and project, and the user adjusts it appropriately.

[0484] 5. Identify waste

[0485] The server quickly identifies wasteful costs and excessive spending based on outliers, which are then evaluated by AI algorithms to highlight areas for improvement.

[0486] Example: A server identifies unnecessarily high communication costs and analyzes the cause.

[0487] 6. Cost-cutting measures

[0488] The server generates specific cost-cutting proposals based on the results of waste identification, and also simulates the effects of the proposals in advance to set priorities.

[0489] Example: The server proposes specific measures for optimizing energy usage and prioritizes them based on their importance.

[0490] 7. Implementation Plan

[0491] The server then creates a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments.

[0492] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[0493] 8. Cost reduction implementation

[0494] The terminal carries out the execution activities based on the formulated plan, and the server monitors this in real time and records the progress.

[0495] Example: A user implements new energy usage guidelines and a server monitors their progress.

[0496] 9. Confirmation of effectiveness

[0497] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0498] Example: A server reviews the effectiveness of new energy usage methods and suggests further improvements.

[0499] Each step of this system automates a company's cost management and improves operational efficiency by accurately and quickly identifying and eliminating waste.

[0500] The processing flow will be explained below.

[0501] Step 1:

[0502] The server accesses a data source, such as an ERP system or accounting software, and sends an API request to retrieve expense data, which includes verifying credentials and specifying the time period to retrieve.

[0503] Example: A server accesses a company's ERP system and downloads last month's expense data using an API.

[0504] Step 2:

[0505] The server checks the integrity and completeness of the retrieved data, adapting to the data format and content and identifying missing or outlier values.

[0506] Example: Standardizing the date format and number of digits in data collected by the server and detecting inappropriate values.

[0507] Step 3:

[0508] The server uses statistical models and predictive algorithms to impute missing values, and also standardizes and cleans the data.

[0509] Example: A server uses a predictive model to fill in missing energy usage data and correct improper formatting.

[0510] Step 4:

[0511] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[0512] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[0513] Step 5:

[0514] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[0515] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[0516] Step 6:

[0517] The server automatically categorizes expense data by category, allowing users to perform detailed analysis by specific category and sorting the data.

[0518] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[0519] Step 7:

[0520] The server identifies wasteful costs using an outlier detection algorithm, quickly uncovering expenditure items that fall outside of normal ranges.

[0521] Example: A server detects extremely high communication costs and analyzes the background.

[0522] Step 8:

[0523] The server generates specific cost-cutting proposals based on the identified waste, and the AI ​​automatically devise effective improvement methods and provides them to the user.

[0524] Example: Propose specific measures to improve the energy efficiency of servers.

[0525] Step 9:

[0526] The server assigns a priority to each suggestion, allowing the user to execute the most effective suggestion first.

[0527] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[0528] Step 10:

[0529] The server creates an action plan based on the reduction proposals, generates a plan including detailed procedures and schedules, and presents it to the user.

[0530] Example: The server creates a specific action plan for reducing energy usage, which the user confirms.

[0531] Step 11:

[0532] The terminal executes cost-cutting activities based on the established plan, and the server monitors this process in real time and records the progress.

[0533] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[0534] Step 12:

[0535] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0536] Example: A server reviews the effectiveness of new energy usage methods and suggests further improvements.

[0537] As described above, by performing specific operations at each step, a system is provided that realizes cost optimization for a company.

[0538] Example 1

[0539] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0540] Current corporate cost management systems often involve manual data collection and analysis processes, resulting in a lack of efficiency. It also makes it difficult to detect wasteful costs and outliers early on, making it difficult to propose effective cost-cutting measures. Furthermore, the consistency and completeness of the collected data cannot be guaranteed, making it difficult to obtain reliable analysis results. To solve these problems and improve corporate operational efficiency, automated data collection, cleansing, and analysis processes are needed.

[0541] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0542] In this invention, the server includes means for automatically collecting expense data from data sources, means for verifying the consistency and completeness of the collected expense data and imputing missing values ​​using a predictive model, means for performing statistical processing on the collected and imputed data to identify outliers, means for generating cost reduction proposals using an AI algorithm based on the identified outliers, means for simulating the effects of the generated proposals in advance and setting priorities, and means for formulating action plans based on the priorities and monitoring their progress and effectiveness in real time, thereby enabling companies to efficiently collect and analyze data and quickly identify and reduce wasteful costs.

[0543] "Data source" refers to the systems and software that provide a company's expense data.

[0544] "Expense data" refers to data that includes information about expenses incurred by a company in various activities.

[0545] "Integrity" refers to a state in which data is consistent and free of contradictions.

[0546] "Complete" refers to the state in which data contains all necessary information.

[0547] A "predictive model" refers to an algorithm or mathematical model that estimates future values ​​based on trends and patterns in data.

[0548] "Statistical processing" refers to methods for summarizing and characterizing data as part of data analysis.

[0549] An "outlier" is a value in the data that is significantly different from the other values.

[0550] "AI algorithm" refers to a data processing method based on machine learning and artificial intelligence.

[0551] "Cost reduction proposals" refer to specific methods and measures for reducing unnecessary costs.

[0552] "Simulation" refers to the process of testing the effectiveness of a proposal in a virtual environment.

[0553] "Priority" refers to ranking multiple items or proposals according to their importance.

[0554] "Implementation plan" refers to a plan for specifically implementing the proposed cost reduction measures.

[0555] "Real-time" refers to a state in which processing and data collection occur immediately.

[0556] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[0557] The main hardware of the system is a server, which automatically collects expense data from data sources such as a company's ERP system and accounting software. Specifically, it retrieves the necessary data from these sources by sending API requests. The server also checks the consistency and completeness of the collected data and imputes missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The server also standardizes the data format and corrects inappropriate values.

[0558] The server then performs statistical operations on the organized data to identify outliers. For example, it uses Pandas to calculate the standard deviation of the data frame to identify outliers. The server also performs trend analysis on the historical data and uses forecasting models such as ARIMA models to forecast future costs.

[0559] The server then automatically categorizes expense data by department and project, allowing users to manually adjust this as needed. It uses Python regular expressions to parse expense record descriptions and classify them into the appropriate categories. Users can review and modify the categorization results using a web interface.

[0560] To identify wasteful costs, the server uses AI algorithms (e.g., random forests) to assess wasteful costs and excessive spending based on outliers, allowing for quick identification of areas for improvement.

[0561] The server also generates specific cost-reduction proposals based on the results of identifying waste. The generated proposals have the ability to simulate their effects in advance and set priorities. For example, the server generates a proposal for installing new lighting equipment and simulates its energy-saving effects. It also calculates the expected cost savings for each proposal and creates a priority list.

[0562] Furthermore, the server creates an action plan based on the priority, which the user can review and ultimately adjust.Specific action plans for each department are created in Excel files, which the user can review and modify via a web interface.

[0563] Finally, the device carries out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. For example, a user implements new energy usage guidelines, and the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard. After this implementation, the server re-analyzes the post-improvement data to confirm the effectiveness of the implementation of the new energy usage guidelines.

[0564] Example prompts to input to a generative AI model:

[0565] "Please explain a program that collects expense data from an ERP system, cleanses the data by imputing missing values ​​with a predictive model, identifies outliers through statistical processing and trend analysis, generates optimization proposals, and formulates specific implementation plans."

[0566] Through this system, companies can streamline cost management and reduce wasteful spending.

[0567] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0568] Step 1:

[0569] Data collection

[0570] The server automatically collects expense data from ERP systems and accounting software. Specifically, the server sends API requests to retrieve the required data from these data sources. The input is the API endpoint of each data source, and the output is the collected expense data. For example, the server sends a request such as "GET / api / expenses?month=2023-09" to the ERP system, receives expense data as a response, and stores it in the database.

[0571] Step 2:

[0572] Data Cleansing

[0573] The server checks the consistency and completeness of the collected expense data and imputes missing values. Specifically, it predicts missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The input is the collected raw expense data, and the output is cleansed data with missing values ​​imputed and a unified format. The server runs a script to unify the date format to "YYYY-MM-DD" and corrects inappropriate values.

[0574] Step 3:

[0575] Cost Analysis

[0576] The server performs statistical processing on the cleansed data to identify outliers. The input is the cleansed expense data, and the output is an analysis showing outliers and trends. For example, the server uses Pandas to calculate the standard deviation of the data frame to identify outliers. It also uses an ARIMA model to forecast future cost trends using data from the past 12 months.

[0577] Step 4:

[0578] Category Classification

[0579] The server categorizes expense data by department or project, allowing users to manually adjust as needed. The input is the analyzed expense data, and the output is the categorized expense data. The server uses Python regular expressions to parse the expense record description and classify it into the appropriate category. Users can view the categorization results through a web interface and make manual adjustments.

[0580] Step 5:

[0581] Identifying waste

[0582] The server identifies wasteful costs and excessive expenditures based on outliers. The input is the analysis results, including outliers, and the output is a list of wasteful costs. Specifically, the server uses an AI algorithm (e.g., random forest) to evaluate wasteful expenditures. The server detects abnormally high communication costs and displays them in a histogram.

[0583] Step 6:

[0584] cost-cutting measures

[0585] The server generates cost-saving proposals based on the results of waste identification. The input is a list of wasteful costs, and the output is specific cost-saving proposals. The server generates proposals for installing new lighting equipment and simulates their energy-saving effects. It calculates the expected cost savings for each proposal and creates a priority list.

[0586] Step 7:

[0587] Implementation Plan

[0588] The server creates an action plan for cost-cutting proposals, which the user can review and adjust. The input is the cost-cutting proposals, and the output is a specific implementation plan. For example, the server creates a specific action plan for each department in an Excel file, and the user can review the plan via a web interface and modify it as needed.

[0589] Step 8:

[0590] Cost reduction implementation

[0591] The terminals carry out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. The input is the implementation plan, and the output is progress data of the implementation activities. For example, when a user puts new energy usage guidelines into practice, the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard.

[0592] Step 9:

[0593] Confirmation of effectiveness

[0594] The server re-analyzes the post-implementation data and confirms the effects of the improvements. The input is the post-implementation data, and the output is the analysis of the effects. The server analyzes the data after the new energy usage guidelines are implemented and displays the effects in a graph. The server then presents the user with a report showing the potential for further cost reductions.

[0595] Through this series of steps, companies can streamline cost management and quickly identify and reduce wasteful spending.

[0596] (Application example 1)

[0597] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0598] Corporate cost management is complex, time-consuming, and labor-intensive, often resulting in a lot of unnecessary expenses. Furthermore, logistics centers often lack real-time information on resource usage, making efficient cost reduction difficult. The present invention aims to solve these problems by providing a system for efficiently managing corporate expenses and reducing unnecessary costs.

[0599] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0600] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost reduction proposals based on the identified outliers, means for formulating plans for implementing the generated proposals and monitoring their progress and effectiveness, and means for understanding the resource usage status of the logistics center in real time via a smartphone application and quickly identifying and reducing unnecessary costs. This automates corporate expense management and enables real-time cost management and reduction at logistics centers.

[0601] A "data source" is an information system or database that provides a company's expense data.

[0602] "Expense data" refers to data that includes all expense information related to the day-to-day running of a company.

[0603] "Integrity" refers to a state in which data is consistent and free of contradictions.

[0604] "Complete" refers to the state in which all data is present without any missing data.

[0605] "Completion" refers to the process of filling in missing data using predictive models, etc.

[0606] An "outlier" is a value that is significantly different from the rest of the data and is outside the range of what is normally expected.

[0607] A "cost reduction proposal" is a specific measure or action plan that should be implemented to reduce wasteful expenses.

[0608] "Plan" refers to the design of specific steps and actions to implement a proposal.

[0609] "Progress" is a state or condition that indicates how far a planned action has been carried out.

[0610] "Effectiveness" is the result that shows how much cost savings the implemented proposals and plans actually achieved.

[0611] A "smartphone application" is a software program that runs on a smartphone and acts as a user interface to provide real-time information and enable operation.

[0612] A "logistics center" is a base for storing, managing, and delivering goods, and is a facility that carries out efficient logistics activities.

[0613] "Resource usage" refers to the utilization of energy, labor, equipment, etc. within a logistics center.

[0614] This invention relates to a system for efficiently managing corporate costs and resource usage at a logistics center. This system can be realized using a smartphone application and a server.

[0615] 1. System Configuration

[0616] The server automatically collects expense data from data sources and checks the consistency and completeness of the collected expense data. It supplements missing data as needed and analyzes the collected data to identify outliers. It generates cost-cutting proposals based on the outliers, develops detailed plans for implementing the proposals, and monitors their progress and effectiveness. Furthermore, a smartphone application allows users to grasp the distribution center's resource usage in real time, quickly identifying and reducing unnecessary costs.

[0617] 2. Hardware and Software

[0618] The server uses the following hardware and software:

[0619] Hardware: A server with a powerful processor and ample storage capacity

[0620] Software: ERP system and accounting software APIs for data collection, pandas and NumPy for data cleansing, TensorFlow and Scikit-learn for statistical processing and outlier detection, Flask / Django as web application frameworks, Firebase / Firestore as real-time databases

[0621] The smartphone application has the following features:

[0622] Data collection: Collect data from ERP systems, sensors in distribution centers, and barcode readers.

[0623] Real-time monitoring: View resource usage in your distribution center in real time

[0624] Notification function: Alerts when unnecessary costs or abnormal values ​​occur

[0625] 3. Program Processing Overview

[0626] The server cleanses expense data collected from data sources for consistency and completeness, and uses predictive models to fill in any missing data. It then performs basic statistical processing on the cleansed data to identify outliers. Based on the identified outliers, an AI algorithm generates cost-saving proposals and performs simulations to verify their effectiveness. Finally, it develops prioritized cost-saving plans, and monitors their progress and effectiveness in real time.

[0627] 4. Examples and prompts

[0628] For example, if you want to predict consumption trends for the next year based on energy consumption data from a logistics center over the past 12 months, you can input the following prompt into the generative AI model:

[0629] "Given the last 12 months of energy consumption data for a distribution center, please generate a model that will predict consumption trends for the next fiscal year. Also, please generate and prioritize specific proposals for cost reduction."

[0630] Using this prompt, the AI ​​model generates output through the following steps:

[0631] 1. Preprocessing of input data

[0632] 2. Trend Forecasting

[0633] 3. Outlier detection

[0634] 4. Cost reduction proposals

[0635] 5. Prioritization

[0636] This system will significantly improve the efficiency of cost management at logistics centers and enable real-time cost reductions.

[0637] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0638] Step 1:

[0639] Data collection

[0640] The server collects expense and resource usage data from the ERP system, sensors in the distribution center, and barcode readers. As input, it sends API requests to retrieve the required data. For example, it accesses the ERP system to download expense reports for the past 12 months and retrieves real-time energy consumption data from sensors. As output, these data are passed to subsequent processing steps.

[0641] Step 2:

[0642] Data Cleansing

[0643] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes the date format of the data, and corrects improper values. It receives the collected raw data as input and produces cleansed and organized data as output.

[0644] Step 3:

[0645] Cost Analysis

[0646] The server performs basic statistical processing on the cleaned data to identify outliers. It also performs trend analysis on past data to predict future costs. It takes the cleansed data as input and provides outliers and predictions of future cost trends as output. For example, it analyzes energy consumption data from the past 12 months to predict consumption trends for the next year.

[0647] Step 4:

[0648] Category Classification

[0649] The server automatically categorizes expense data and allows users to analyze it by specific categories. Users can also manually adjust the categorization results as needed. It uses the analyzed data as input and generates categorized data as output. For example, categorizing expense data by department or project.

[0650] Step 5:

[0651] Identifying waste

[0652] The server quickly identifies wasteful costs and excessive expenditures based on outliers. AI algorithms evaluate these and identify areas for improvement. It receives outlier data as input and provides the results of wasteful cost identification as output. For example, it identifies unnecessarily high communication costs and analyzes the causes.

[0653] Step 6:

[0654] cost-cutting measures

[0655] The server generates specific cost-saving proposals based on the waste identification results. It also simulates the effectiveness of the proposals in advance and sets priorities. It uses the waste identification results and related data as input and provides cost-saving proposals and their predicted effects as output. For example, it proposes specific measures for optimizing energy use and prioritizes the proposals based on their importance.

[0656] Step 7:

[0657] Implementation Plan

[0658] The server then develops a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments. It uses the cost-saving proposals as input and provides a detailed implementation plan as output. For example, an energy reduction plan can be developed, which the user can review and adjust to fit their office hours.

[0659] Step 8:

[0660] Cost reduction implementation

[0661] The device performs the execution activities based on the formulated plan. The server monitors this in real time and records the progress. It uses the implementation plan as input and generates execution progress data as output. For example, a user implements new energy usage guidelines and the server monitors their progress.

[0662] Step 9:

[0663] Confirmation of effectiveness

[0664] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements. If necessary, it will propose further optimizations. It uses the execution result data as input and provides the results of the effectiveness check and further improvement suggestions as output. For example, it can confirm the effectiveness of a new energy usage method and propose further improvements.

[0665] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0666] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management. In particular, it is characterized by combining an emotion engine that recognizes the user's emotions and providing customized suggestions according to the user's emotions.

[0667] The system of the present invention includes the following elements:

[0668] 1. Data Collection

[0669] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data.

[0670] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[0671] 2. Data cleansing

[0672] The server checks the consistency and completeness of the retrieved data, imputes missing values ​​using predictive algorithms, and standardizes the format.

[0673] Example: The server corrects mismatched date formats and missing values.

[0674] 3. Cost Analysis

[0675] The server performs statistical processing to identify outliers in expenses and then performs trend analysis to predict future costs.

[0676] Example: A server analyzes energy consumption data and predicts consumption trends for the next year.

[0677] 4. Categorization

[0678] The server automatically categorizes expense data and allows users to review and adjust the details.

[0679] Example: A server categorizes expenses by department or project.

[0680] 5. Identify waste

[0681] The server uses an anomaly detection algorithm to quickly identify wasteful costs, and AI evaluates the potential for improvement.

[0682] Example: A server identifies high communication charges and analyzes their causes.

[0683] 6. Cost reduction proposals

[0684] The server generates specific cost-cutting proposals based on the identified waste, simulates the effectiveness of the proposals, and sets priorities.

[0685] Example: Propose specific measures to improve the energy efficiency of a server and list them in order of importance.

[0686] Additionally, the system of the present invention incorporating the emotion engine includes the following elements:

[0687] 1. Emotion recognition

[0688] The server uses an emotion engine to recognize the user's emotions and analyzes the user's writings and comments to identify the user's emotional state.

[0689] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[0690] 2. Emotion-based personalized recommendations

[0691] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. For example, if the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[0692] Example: When a user is feeling stressed, the server prioritizes suggestions for reducing stress that are relatively easy to implement.

[0693] 3. Emotional Feedback

[0694] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[0695] Example: The server displays a dashboard showing the user's current emotional state and provides suggestions for improvement based on that.

[0696] For example, if a company's human resources manager is feeling stressed, the system will recognize that emotion and first offer relatively simple and effective improvement suggestions, allowing the manager to start with suggestions that are easy to implement. The system also monitors the effectiveness of the suggestions after implementation and, if further suggestions are needed, suggests next steps based on the manager's emotional state.

[0697] In this way, the present invention aims to make corporate cost management more effective and user-friendly by combining AI technology and emotion recognition technology.

[0698] The processing flow will be explained below.

[0699] Step 1:

[0700] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data, which includes verifying credentials and specifying the time period to retrieve.

[0701] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[0702] Step 2:

[0703] The server checks the integrity and completeness of the retrieved data, adapts to the data format and content, identifies missing or outlier values, and imputes data as needed.

[0704] Example: Standardizing the date format and number of digits in data collected by the server, and detecting and correcting inappropriate values.

[0705] Step 3:

[0706] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[0707] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[0708] Step 4:

[0709] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[0710] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[0711] Step 5:

[0712] The server automatically categorizes expense data by category, allows users to perform detailed analysis for each specific category, and allows users to manually adjust the categorization results as needed.

[0713] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[0714] Step 6:

[0715] The server uses AI technology to identify outliers, using predictive algorithms to spot spending items that fall outside of the normal range.

[0716] Example: A server detects extremely high communication costs and analyzes the background.

[0717] Step 7:

[0718] The server generates specific cost-saving proposals based on the identified outliers, simulates the effects of the proposals, and provides the proposals in an easy-to-implement format for the user.

[0719] Example: Propose specific measures to improve the energy efficiency of servers.

[0720] Step 8:

[0721] The server prioritizes the generated reduction proposals, allowing the user to implement the most effective proposals first.

[0722] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[0723] Step 9:

[0724] The server creates a detailed implementation plan based on the reduction proposals, including specific steps and schedules, which the user can review and adjust.

[0725] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[0726] Step 10:

[0727] The terminal executes cost-cutting activities based on the established plan, and the server monitors the progress in real time and provides feedback to the user.

[0728] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[0729] Step 11:

[0730] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0731] Example: A server reviews the effectiveness of new energy usage methods and suggests further improvements.

[0732] (A system incorporating an emotion engine)

[0733] Step 12:

[0734] The server uses an emotion engine to recognize the user's emotions by analyzing the user's posts and comments to identify their emotional state.

[0735] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[0736] Step 13:

[0737] The server customizes the content and priorities of cost-saving proposals based on the user's recognized emotions, giving priority to proposals that are less burdensome, taking into account the user's emotional state.

[0738] Example: When a server detects a user is feeling stressed, it prioritizes suggestions for reducing stress that are relatively easy to implement.

[0739] Step 14:

[0740] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[0741] Example: The server displays a dashboard showing the user's current emotional state and provides appropriate improvement suggestions based on that.

[0742] In this way, a system that combines an emotion engine can make flexible cost optimization proposals that take into account the user's emotional state, aiming to make corporate cost management more efficient and user-friendly.

[0743] Example 2

[0744] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0745] Conventional cost management systems collect and analyze expense data, generate cost-reduction proposals, and formulate implementation plans, but they are unable to provide effective proposals that take into account the user's emotional state. This can lead to stress and difficulty in implementing the proposals. Therefore, there is a need for a system that recognizes the user's emotional state and provides customized proposals based on that state.

[0746] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0747] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost-cutting proposals based on the identified outliers, means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness, means for identifying the user's emotional state using an emotion engine that recognizes the user's emotions, means for customizing the content and priorities of the cost-cutting proposals based on the user's emotional state, and means for visually providing feedback on the user's emotional state, thereby enabling effective and less burdensome cost-cutting proposals to be made in accordance with the user's emotional state.

[0748] "Data sources" refers to the systems and applications that store a company's expense data, including ERP systems and accounting software.

[0749] "Expense data" is data that contains detailed information about the costs a company incurs on a day-to-day basis.

[0750] "Integrity" means that data is accurate, consistent, and does not contradict each other.

[0751] "Complete" means that the data contains all necessary information and is free of missing parts.

[0752] "Imputation" refers to filling in missing information in data using prediction or imputation algorithms.

[0753] An "outlier" is a value that is extremely outlier compared to other values ​​in a data set, and often includes errors or exceptions.

[0754] A "cost reduction proposal" refers to a proposal that outlines specific action items and strategies for reducing a company's expenses.

[0755] "Plan" refers to a document that contains specific steps and timelines for implementing the generated cost reduction proposals.

[0756] "Progress" refers to the status of the implementation of planned specific actions, and indicates how much progress has been made over time.

[0757] "Effects" refer to the results obtained as a result of implementing a plan, and specifically include the degree of cost reduction and improved efficiency.

[0758] "Emotion engine" refers to software equipped with artificial intelligence technology to recognize and analyze a user's emotional state.

[0759] "Emotional state" refers to the psychological state such as stress or satisfaction felt by the user.

[0760] "Feedback" refers to providing the analysis results and suggestions to the user visually or in other ways, allowing the user to check their own status and the suggestions.

[0761] This invention is a system that utilizes AI technology to improve the efficiency of corporate cost management, and is characterized by its incorporation of an emotion engine that recognizes user emotions. To implement this system, hardware and software including the following elements are used.

[0762] 1. Data Collection

[0763] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data. Specifically, the server sends an HTTP request and receives data in JSON format from the ERP system or accounting software.

[0764] Example: A server sends a request to "https: / / api.erp.example.com / v1 / expenses?month=last" to retrieve last month's expense data from a company's ERP system.

[0765] 2. Data cleansing

[0766] The server checks the integrity and completeness of the data retrieved, imputes missing values ​​using predictive algorithms, and standardizes formats, including detecting duplicate entries in the database and using a reverse dictionary to infer missing dates.

[0767] Example: A server corrects inconsistencies in date formats between expense data entries and imputes missing date fields with a predictive algorithm.

[0768] 3. Cost Analysis

[0769] The server performs statistical processing to identify outliers in expenses and forecast future costs. Specifically, it uses the Pandas library to detect outliers and the Prophet model to forecast future cost trends.

[0770] Example: A server analyzes energy consumption data and predicts energy consumption trends for the next year.

[0771] 4. Categorization

[0772] The server automatically categorizes expense data into categories and allows users to review and adjust the details, including using machine learning models to categorize data into specific categories.

[0773] Example: A server organizes expense data into categories such as communication expenses, travel expenses, and advertising expenses.

[0774] 5. Identify waste

[0775] The server uses anomaly detection algorithms to identify wasteful costs and evaluate potential areas for improvement, using techniques such as Isolation Forest and One-Class SVM.

[0776] Example: The server identifies communication expenses that are significantly higher than average as an anomaly and analyzes the details.

[0777] 6. Cost reduction proposals

[0778] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[0779] Example: A server generates a list of proposals for replacing lighting with LEDs to improve energy efficiency, based on estimated results.

[0780] 7. Emotion recognition

[0781] The server uses an emotion engine to recognize user emotions and identify the emotional state from user posts and statements, using a natural language processing model (e.g., BERT).

[0782] Example: A server analyzes a user's message "Work has been very stressful lately" and determines that the user is feeling stressed.

[0783] 8. Emotion-based personalized recommendations

[0784] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. If the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[0785] Example: When a user feels stressed, the server displays a low-impact reduction suggestion such as "reduce the frequency of meetings."

[0786] 9. Emotional Feedback

[0787] The server provides visual feedback of the user's emotional state, allowing the user to check the state and select suggestions using a visualization tool.

[0788] Example: The server displays the user's emotional state on a dashboard as "Stress level: High" and provides suggestions for improvement based on that.

[0789] This system allows companies to achieve efficient and user-friendly cost management, allowing users to receive cost-saving suggestions that take their emotional state into account and are easy to implement.

[0790] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0791] Program processing flow

[0792] Step 1: Data collection

[0793] The server contacts the data source and sends an API request to retrieve expense data.

[0794] Input: API endpoint URL and authentication information configured on the server.

[0795] Specific behavior: The server sends an HTTP request to an endpoint such as "https: / / api.erp.example.com / v1 / expenses?month=last".

[0796] Output: Expense data in JSON format returned from your ERP system or accounting software.

[0797] Step 2: Data cleansing

[0798] The server checks the consistency and completeness of the data it retrieves, fills in missing values, and standardizes the format.

[0799] Input: Captured expense data in JSON format.

[0800] What it does: The server uses a database library to read data, detects duplicate entries, fills in missing data with a predictive algorithm, and standardizes data formats (for example, standardizing date formats to "YYYY-MM-DD").

[0801] Output: A clean dataset that has been checked for consistency and completeness, imputed for missing values, and has a uniform format.

[0802] Step 3: Cost analysis

[0803] The server analyzes the cleansed data to identify outliers and predict future costs.

[0804] Input: The cleansed dataset.

[0805] What it does: The server uses a statistical library to detect outliers in the data and a time series forecasting model (e.g., Prophet) to predict future cost trends.

[0806] Output: Identified outliers and future cost forecast data.

[0807] Step 4: Categorization

[0808] The server automatically categorizes expense data and allows users to review and adjust the details.

[0809] Input: The cleansed dataset.

[0810] What it does: The server uses machine learning models to categorize expense data by category (e.g., communication, travel, advertising).

[0811] Output: Expense data broken down by category.

[0812] Step 5: Identify waste

[0813] The server uses an anomaly detection algorithm to identify wasteful costs and evaluate potential for improvement.

[0814] Input: Expense data broken down by category.

[0815] How it works: The server uses outlier detection algorithms such as Isolation Forest and One-Class SVM to identify unreasonably high expenditures.

[0816] Output: Identified waste costs and a detailed report.

[0817] Step 6: Cost reduction proposal

[0818] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[0819] Input: Detailed data on waste costs.

[0820] Specific operation: The server generates improvement measures using heuristic rules and verifies their effectiveness using a simulation model.

[0821] Output: Prioritized cost-saving proposals and their predicted effectiveness.

[0822] Step 7: Emotion Recognition

[0823] The server uses an emotion engine to recognize the user's emotions and analyzes the user's emotional state from their statements and writings.

[0824] Input: User message and feedback data.

[0825] What happens: The server analyzes the text data using a natural language processing model (e.g., BERT) to identify the user's emotional state.

[0826] Output: User's emotional state report.

[0827] Step 8: Customize your suggestions based on emotions

[0828] The server customizes the suggestions and priorities based on the user's emotions recognized.

[0829] Input: User emotional state report, cost reduction suggestions.

[0830] Specific operation: The server selects and customizes suggestions that are easy to implement and have high priority based on the user's emotional state.

[0831] Output: Customized cost-saving proposals.

[0832] Step 9: Emotional Feedback

[0833] The server visually feeds back the user's emotional state, allowing the user to confirm the state and select suggestions.

[0834] Input: customized cost-saving suggestions, user emotional state report.

[0835] Specific operation: The server uses visualization tools such as a dashboard to display the user's emotional state and suggestions.

[0836] Output: Emotional feedback and improvement suggestions on a visualized dashboard.

[0837] (Application example 2)

[0838] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0839] Conventional cost management systems collect and analyze expense data, identify outliers, and generate cost-reduction proposals. However, these processes do not take into account the user's emotional state, often resulting in stress. Furthermore, in logistics centers, worker comfort and efficiency are important, and workload adjustments based on individual emotional states are required. However, current systems lack an effective means of doing this, resulting in insufficient improvements in work efficiency and reductions in worker stress.

[0840] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost reduction proposals based on the identified outliers, means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness, means for determining the user's emotional state using emotion recognition technology and customizing the content and priority of the proposals based on the determined emotional state, and means having a human interface for presenting the customized proposals to the user. This enables efficient cost management that takes the user's emotional state into consideration, reducing worker stress and improving work efficiency at logistics centers.

[0841] "Data sources" refers to information sources such as ERP systems or accounting software that store a company's expense data.

[0842] "Expense data" is detailed information about the costs a company incurs on a daily basis, including travel, communication, and energy costs.

[0843] "Integrity" refers to data that is consistently accurate and consistent across different data sets.

[0844] "Complete" refers to the state in which all data is available and there are no missing values ​​or incomplete information.

[0845] "Imputation" refers to the process of estimating and adding missing data to complete a dataset.

[0846] An "outlier" is a data value that deviates from normal patterns or acceptable ranges.

[0847] "Cost reduction proposals" refer to specific action plans and measures to reduce a company's expenses.

[0848] A "plan" is a schedule that defines the steps and resources needed to achieve a specific goal.

[0849] "Progress" refers to the status of execution and degree of success of a plan or project.

[0850] "Effect" refers to the results or outcomes that result from a certain action or measure.

[0851] "Emotion recognition technology" refers to technology that identifies a user's emotional state (e.g., stress, joy, anger, etc.) from facial expressions and vocal tone.

[0852] "User" refers to the person who operates the system and manages and analyzes expense data.

[0853] "Emotional state" refers to a user's current psychological state as determined by emotion recognition technology.

[0854] "Human interface" refers to the interface through which a user interacts with a system and exchanges information.

[0855] This invention provides a system aimed at efficient cost management and worker stress reduction in logistics centers. This system utilizes AI technology and emotion recognition technology to make suggestions based on the user's emotional state. Specific embodiments are described below.

[0856] The system consists of a server and smart glasses as its main components. The server automatically collects expense data from data sources such as a company's ERP system or accounting software, including the means to retrieve the required data using API requests.

[0857] The server checks the consistency and completeness of collected expense data, fills in missing values ​​with a predictive algorithm, and standardizes the format, thereby ensuring data quality.

[0858] The server then performs statistical processing and analyses on the collected and stored data, identifying outliers and conducting trend analysis to predict future costs, which in turn generates cost-saving recommendations.

[0859] The server uses emotion recognition technology to collect data through the smart glasses to determine the worker's emotional state. Specifically, it uses facial recognition and voice analysis to determine the worker's emotional state from their facial expressions and tone. Based on the determined emotional state, the server customizes the content and priorities of suggestions.

[0860] The customized suggestions are sent from the server to the smart glasses and presented to the worker through a human interface, so that suggestions for a lighter workload are displayed preferentially to workers who are feeling stressed.

[0861] The server uses TensorFlow and OpenCV to build emotion recognition models, Flask framework to manage data on the server side, and SQLite to operate a local database. Communication with the smart glasses is via Bluetooth Low Energy (BLE).

[0862] For example, if a worker at a logistics center is feeling stressed, the system will recognize that emotion and suggest lighter tasks first, such as avoiding carrying heavy items and prioritizing stocking lighter items, thereby reducing the worker's stress and allowing them to work efficiently.

[0863] Below is an example of a prompt sentence that uses a generative AI model to design suggestions based on the worker's emotions.

[0864] Example prompt sentence:

[0865] Consider a graphical interface design that uses an emotion engine to identify the emotions of workers and suggest lighter tasks to stressed workers. In particular, be sure to consider the use of colors and layout to reduce stress.

[0866] In this way, this system simultaneously improves the cost performance and work efficiency of logistics centers through efficient management of expense data and suggestions based on the emotional state of workers.

[0867] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0868] Step 1:

[0869] The server collects expense data from data sources, specifically, from ERP systems and accounting software using API requests. The input is the access information of the data source based on the API request, and the output is the collected raw expense data.

[0870] Step 2:

[0871] The server checks the collected expense data for consistency and completeness and imputes it if necessary, correcting inconsistent date formats and imputing missing values ​​using predictive algorithms. The input is the raw expense data collected, and the output is cleaned data that has been checked for consistency and completeness.

[0872] Step 3:

[0873] The server analyzes the collected and imputed data to identify outliers. It performs statistical processing to detect outliers that deviate from the normal range. The input is the cleaned expense data, and the output is a list of outliers.

[0874] Step 4:

[0875] The server generates cost reduction proposals based on the identified outliers. It analyzes the causes of the outliers and proposes specific cost reduction measures accordingly. The input is a list of outliers, and the output is the generated cost reduction proposals.

[0876] Step 5:

[0877] The server uses emotion recognition technology to determine the worker's emotional state. It uses the camera and microphone in the smart glasses to perform facial recognition and voice analysis to identify the worker's emotional state. The input is real-time video and audio data, and the output is the determined emotional state.

[0878] Step 6:

[0879] The server customizes the content and priority of the proposals based on the determined emotional state. If the user is feeling stressed, it prioritizes proposals that are less stressful. The input is the determined emotional state and the generated cost-saving proposals, and the output is the customized proposals.

[0880] Step 7:

[0881] The server sends the customized suggestions to the smart glasses for presentation to the worker. The suggestions are displayed to the worker via a human interface using Bluetooth Low Energy (BLE). The input is the customized suggestions, and the output is the suggestions displayed on the smart glasses display.

[0882] Step 8:

[0883] The server develops an implementation plan for the proposals and monitors their progress and effectiveness. It evaluates the results of implementing the proposed cost-saving measures and obtains feedback. The inputs are the implemented proposals and their results data, and the outputs are progress reports and effectiveness evaluation reports.

[0884] By carrying out the above processing steps, efficient cost management and stress reduction for workers at the logistics center can be achieved.

[0885] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0886] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0887] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0888] [Third embodiment]

[0889] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0890] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0891] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0892] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0893] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0894] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0895] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0896] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0897] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0898] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0899] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0900] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0901] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[0902] The system of the present invention achieves cost optimization mainly through the following steps.

[0903] 1. Data Collection

[0904] The server automatically collects the company's expense data from data sources such as ERP systems, accounting software, etc. It sends API requests to retrieve the required data.

[0905] Example: A server accesses a company's ERP system and downloads last month's expense report.

[0906] 2. Data cleansing

[0907] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes formats, and corrects invalid values.

[0908] Example: Standardizing the date format of data collected by the server and correcting incorrect values.

[0909] 3. Cost Analysis

[0910] The server performs basic statistical processing on the organized data to identify outliers, and also performs trend analysis on past data to predict future costs.

[0911] Example: A server analyzes energy consumption data from the past 12 months and predicts consumption trends for the next year.

[0912] 4. Categorization

[0913] The server automatically categorizes expense data and allows users to analyze it by specific categories, with the option to manually adjust the categorization as needed.

[0914] Example: The server categorizes collected expense data by department and project, and the user adjusts it appropriately.

[0915] 5. Identify waste

[0916] The server quickly identifies wasteful costs and excessive spending based on outliers, which are then evaluated by AI algorithms to highlight areas for improvement.

[0917] Example: A server identifies unnecessarily high communication costs and analyzes the cause.

[0918] 6. Cost-cutting measures

[0919] The server generates specific cost-cutting proposals based on the results of waste identification, and also simulates the effects of the proposals in advance to set priorities.

[0920] Example: The server proposes specific measures for optimizing energy usage and prioritizes them based on their importance.

[0921] 7. Implementation Plan

[0922] The server then creates a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments.

[0923] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[0924] 8. Cost reduction implementation

[0925] The terminal carries out the execution activities based on the formulated plan, and the server monitors this in real time and records the progress.

[0926] Example: A user implements new energy usage guidelines and a server monitors their progress.

[0927] 9. Confirmation of effectiveness

[0928] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0929] Example: A server checks the effectiveness of new energy usage methods and suggests further improvements.

[0930] Each step of this system automates a company's cost management and improves operational efficiency by accurately and quickly identifying and eliminating waste.

[0931] The processing flow will be explained below.

[0932] Step 1:

[0933] The server accesses a data source, such as an ERP system or accounting software, and sends an API request to retrieve expense data, which includes verifying credentials and specifying the time period to retrieve.

[0934] Example: A server accesses a company's ERP system and downloads last month's expense data using an API.

[0935] Step 2:

[0936] The server checks the integrity and completeness of the retrieved data, adapting to the data format and content and identifying missing or outlier values.

[0937] Example: Standardizing the date format and number of digits in data collected by the server and detecting inappropriate values.

[0938] Step 3:

[0939] The server uses statistical models and predictive algorithms to impute missing values, and also standardizes and cleans the data.

[0940] Example: A server uses a predictive model to fill in missing energy usage data and correct improper formatting.

[0941] Step 4:

[0942] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[0943] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[0944] Step 5:

[0945] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[0946] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[0947] Step 6:

[0948] The server automatically categorizes expense data by category, allowing users to perform detailed analysis by specific category and sorting the data.

[0949] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[0950] Step 7:

[0951] The server identifies wasteful costs using an outlier detection algorithm, quickly uncovering expenditure items that fall outside of normal ranges.

[0952] Example: A server detects extremely high communication costs and analyzes the background.

[0953] Step 8:

[0954] The server generates specific cost-cutting proposals based on the identified waste, and the AI ​​automatically devise effective improvement methods and provides them to the user.

[0955] Example: Propose specific measures to improve the energy efficiency of servers.

[0956] Step 9:

[0957] The server assigns a priority to each suggestion, allowing the user to execute the most effective suggestion first.

[0958] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[0959] Step 10:

[0960] The server creates an action plan based on the reduction proposals, generates a plan including detailed procedures and schedules, and presents it to the user.

[0961] Example: The server creates a specific action plan for reducing energy usage, which the user confirms.

[0962] Step 11:

[0963] The terminal executes cost-cutting activities based on the established plan, and the server monitors this process in real time and records the progress.

[0964] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[0965] Step 12:

[0966] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[0967] Example: A server checks the effectiveness of new energy usage methods and suggests further improvements.

[0968] As described above, by performing specific operations at each step, a system is provided that realizes cost optimization for a company.

[0969] Example 1

[0970] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0971] Current corporate cost management systems often involve manual data collection and analysis processes, resulting in a lack of efficiency. It also makes it difficult to detect wasteful costs and outliers early on, making it difficult to propose effective cost-cutting measures. Furthermore, the consistency and completeness of the collected data cannot be guaranteed, making it difficult to obtain reliable analysis results. To solve these problems and improve corporate operational efficiency, automated data collection, cleansing, and analysis processes are needed.

[0972] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0973] In this invention, the server includes means for automatically collecting expense data from data sources, means for verifying the consistency and completeness of the collected expense data and imputing missing values ​​using a predictive model, means for performing statistical processing on the collected and imputed data to identify outliers, means for generating cost reduction proposals using an AI algorithm based on the identified outliers, means for simulating the effects of the generated proposals in advance and setting priorities, and means for formulating action plans based on the priorities and monitoring their progress and effectiveness in real time, thereby enabling companies to efficiently collect and analyze data and quickly identify and reduce wasteful costs.

[0974] "Data source" refers to the systems and software that provide a company's expense data.

[0975] "Expense data" refers to data that includes information about expenses incurred by a company in various activities.

[0976] "Integrity" refers to a state in which data is consistent and free of contradictions.

[0977] "Complete" refers to the state in which data contains all necessary information.

[0978] A "predictive model" refers to an algorithm or mathematical model that estimates future values ​​based on trends and patterns in data.

[0979] "Statistical processing" refers to methods for summarizing and characterizing data as part of data analysis.

[0980] An "outlier" is a value in the data that is significantly different from the other values.

[0981] "AI algorithm" refers to a data processing method based on machine learning and artificial intelligence.

[0982] "Cost reduction proposals" refer to specific methods and measures for reducing unnecessary costs.

[0983] "Simulation" refers to the process of testing the effectiveness of a proposal in a virtual environment.

[0984] "Priority" refers to ranking multiple items or proposals according to their importance.

[0985] "Implementation plan" refers to a plan for specifically implementing the proposed cost reduction measures.

[0986] "Real-time" refers to a state in which processing and data collection occur immediately.

[0987] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[0988] The main hardware of the system is a server, which automatically collects expense data from data sources such as a company's ERP system and accounting software. Specifically, it retrieves the necessary data from these sources by sending API requests. The server also checks the consistency and completeness of the collected data and imputes missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The server also standardizes the data format and corrects inappropriate values.

[0989] The server then performs statistical operations on the organized data to identify outliers. For example, it uses Pandas to calculate the standard deviation of the data frame to identify outliers. The server also performs trend analysis on the historical data and uses forecasting models such as ARIMA models to forecast future costs.

[0990] The server then automatically categorizes expense data by department and project, allowing users to manually adjust this as needed. It uses Python regular expressions to parse expense record descriptions and classify them into the appropriate categories. Users can review and modify the categorization results using a web interface.

[0991] To identify wasteful costs, the server uses AI algorithms (e.g., random forests) to assess wasteful costs and excessive spending based on outliers, allowing for quick identification of areas for improvement.

[0992] The server also generates specific cost-reduction proposals based on the results of identifying waste. The generated proposals have the ability to simulate their effects in advance and set priorities. For example, the server generates a proposal for installing new lighting equipment and simulates its energy-saving effects. It also calculates the expected cost savings for each proposal and creates a priority list.

[0993] Furthermore, the server creates an action plan based on the priority, which the user can review and ultimately adjust.Specific action plans for each department are created in Excel files, which the user can review and modify via a web interface.

[0994] Finally, the device carries out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. For example, a user implements new energy usage guidelines, and the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard. After this implementation, the server re-analyzes the post-improvement data to confirm the effectiveness of the implementation of the new energy usage guidelines.

[0995] Example prompts to input to a generative AI model:

[0996] "Please explain a program that collects expense data from an ERP system, cleanses the data by imputing missing values ​​with a predictive model, identifies outliers through statistical processing and trend analysis, generates optimization proposals, and formulates specific implementation plans."

[0997] Through this system, companies can streamline cost management and reduce wasteful spending.

[0998] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0999] Step 1:

[1000] Data collection

[1001] The server automatically collects expense data from ERP systems and accounting software. Specifically, the server sends API requests to retrieve the required data from these data sources. The input is the API endpoint of each data source, and the output is the collected expense data. For example, the server sends a request such as "GET / api / expenses?month=2023-09" to the ERP system, receives expense data as a response, and stores it in the database.

[1002] Step 2:

[1003] Data Cleansing

[1004] The server checks the consistency and completeness of the collected expense data and imputes missing values. Specifically, it predicts missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The input is the collected raw expense data, and the output is cleansed data with missing values ​​imputed and a unified format. The server runs a script to unify the date format to "YYYY-MM-DD" and corrects inappropriate values.

[1005] Step 3:

[1006] Cost Analysis

[1007] The server performs statistical processing on the cleansed data to identify outliers. The input is the cleansed expense data, and the output is an analysis showing outliers and trends. For example, the server uses Pandas to calculate the standard deviation of the data frame to identify outliers. It also uses an ARIMA model to forecast future cost trends using data from the past 12 months.

[1008] Step 4:

[1009] Category Classification

[1010] The server categorizes expense data by department or project, allowing users to manually adjust as needed. The input is the analyzed expense data, and the output is the categorized expense data. The server uses Python regular expressions to parse the expense record description and classify it into the appropriate category. Users can view the categorization results through a web interface and make manual adjustments.

[1011] Step 5:

[1012] Identifying waste

[1013] The server identifies wasteful costs and excessive expenditures based on outliers. The input is the analysis results, including outliers, and the output is a list of wasteful costs. Specifically, the server uses an AI algorithm (e.g., random forest) to evaluate wasteful expenditures. The server detects abnormally high communication costs and displays them in a histogram.

[1014] Step 6:

[1015] cost-cutting measures

[1016] The server generates cost-saving proposals based on the results of waste identification. The input is a list of wasteful costs, and the output is specific cost-saving proposals. The server generates proposals for installing new lighting equipment and simulates their energy-saving effects. It calculates the expected cost savings for each proposal and creates a priority list.

[1017] Step 7:

[1018] Implementation Plan

[1019] The server creates an action plan for cost-cutting proposals, which the user can review and adjust. The input is the cost-cutting proposals, and the output is a specific implementation plan. For example, the server creates a specific action plan for each department in an Excel file, and the user can review the plan via a web interface and modify it as needed.

[1020] Step 8:

[1021] Cost reduction implementation

[1022] The terminals carry out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. The input is the implementation plan, and the output is progress data of the implementation activities. For example, when a user puts new energy usage guidelines into practice, the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard.

[1023] Step 9:

[1024] Confirmation of effectiveness

[1025] The server re-analyzes the post-implementation data and confirms the effects of the improvements. The input is the post-implementation data, and the output is the analysis of the effects. The server analyzes the data after the new energy usage guidelines are implemented and displays the effects in a graph. The server then presents the user with a report showing the potential for further cost reductions.

[1026] Through this series of steps, companies can streamline cost management and quickly identify and reduce wasteful spending.

[1027] (Application example 1)

[1028] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1029] Corporate cost management is complex, time-consuming, and labor-intensive, often resulting in a lot of unnecessary expenses. Furthermore, logistics centers often lack real-time information on resource usage, making efficient cost reduction difficult. The present invention aims to solve these problems by providing a system for efficiently managing corporate expenses and reducing unnecessary costs.

[1030] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1031] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost reduction proposals based on the identified outliers, means for formulating plans for implementing the generated proposals and monitoring their progress and effectiveness, and means for understanding the resource usage status of the logistics center in real time via a smartphone application and quickly identifying and reducing unnecessary costs. This automates corporate expense management and enables real-time cost management and reduction at logistics centers.

[1032] A "data source" is an information system or database that provides a company's expense data.

[1033] "Expense data" refers to data that includes all expense information related to the day-to-day running of a company.

[1034] "Integrity" refers to a state in which data is consistent and free of contradictions.

[1035] "Complete" refers to the state in which all data is present and without any missing data.

[1036] "Completion" refers to the process of filling in missing data using predictive models, etc.

[1037] An "outlier" is a value that is significantly different from the rest of the data and is outside the range of what is normally expected.

[1038] A "cost reduction proposal" is a specific measure or action plan that should be implemented to reduce wasteful expenses.

[1039] "Plan" refers to the design of specific steps and actions to implement a proposal.

[1040] "Progress" is a state or condition that indicates how far a planned action has been carried out.

[1041] "Effectiveness" is the result that shows how much cost savings the implemented proposals and plans actually achieved.

[1042] A "smartphone application" is a software program that runs on a smartphone and acts as a user interface to provide real-time information and enable operation.

[1043] A "logistics center" is a base for storing, managing, and delivering goods, and is a facility that carries out efficient logistics activities.

[1044] "Resource usage" refers to the utilization of energy, labor, equipment, etc. within a logistics center.

[1045] This invention relates to a system for efficiently managing corporate costs and resource usage at a logistics center. This system can be realized using a smartphone application and a server.

[1046] 1. System Configuration

[1047] The server automatically collects expense data from data sources and checks the consistency and completeness of the collected expense data. It supplements missing data as needed and analyzes the collected data to identify outliers. It generates cost-cutting proposals based on the outliers, develops detailed plans for implementing the proposals, and monitors their progress and effectiveness. Furthermore, a smartphone application allows users to grasp the distribution center's resource usage in real time, quickly identifying and reducing unnecessary costs.

[1048] 2. Hardware and Software

[1049] The server uses the following hardware and software:

[1050] Hardware: A server with a powerful processor and ample storage capacity

[1051] Software: ERP system and accounting software APIs for data collection, pandas and NumPy for data cleansing, TensorFlow and Scikit-learn for statistical processing and outlier detection, Flask / Django as web application frameworks, Firebase / Firestore as real-time databases

[1052] The smartphone application has the following features:

[1053] Data collection: Collect data from ERP systems, sensors in distribution centers, and barcode readers.

[1054] Real-time monitoring: View resource usage in your distribution center in real time

[1055] Notification function: Alerts when unnecessary costs or abnormal values ​​occur

[1056] 3. Program Processing Overview

[1057] The server cleanses expense data collected from data sources for consistency and completeness, and uses predictive models to fill in any missing data. It then performs basic statistical processing on the cleansed data to identify outliers. Based on the identified outliers, an AI algorithm generates cost-saving proposals and performs simulations to verify their effectiveness. Finally, it develops prioritized cost-saving plans, and monitors their progress and effectiveness in real time.

[1058] 4. Examples and prompts

[1059] For example, if you want to predict consumption trends for the next year based on a logistics center's energy consumption data from the past 12 months, you can input the following prompt into the generative AI model:

[1060] "Given the last 12 months of energy consumption data for a distribution center, please generate a model that will predict consumption trends for the next fiscal year. Also, please generate and prioritize specific proposals for cost reduction."

[1061] Using this prompt, the AI ​​model generates output through the following steps:

[1062] 1. Preprocessing of input data

[1063] 2. Trend Forecasting

[1064] 3. Outlier detection

[1065] 4. Cost reduction proposals

[1066] 5. Prioritization

[1067] This system will significantly improve the efficiency of cost management at logistics centers and enable real-time cost reductions.

[1068] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1069] Step 1:

[1070] Data collection

[1071] The server collects expense and resource usage data from the ERP system, sensors in the distribution center, and barcode readers. As input, it sends API requests to retrieve the required data. For example, it accesses the ERP system to download expense reports for the past 12 months and retrieves real-time energy consumption data from sensors. As output, these data are passed to subsequent processing steps.

[1072] Step 2:

[1073] Data Cleansing

[1074] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes the date format of the data, and corrects improper values. It receives the collected raw data as input and produces cleansed and organized data as output.

[1075] Step 3:

[1076] Cost Analysis

[1077] The server performs basic statistical processing on the cleaned data to identify outliers. It also performs trend analysis on past data to predict future costs. It takes the cleansed data as input and provides outliers and predictions of future cost trends as output. For example, it analyzes energy consumption data from the past 12 months to predict consumption trends for the next year.

[1078] Step 4:

[1079] Category Classification

[1080] The server automatically categorizes expense data and allows users to analyze it by specific categories. Users can also manually adjust the categorization results as needed. It uses the analyzed data as input and generates categorized data as output. For example, categorizing expense data by department or project.

[1081] Step 5:

[1082] Identifying waste

[1083] The server quickly identifies wasteful costs and excessive expenditures based on outliers. AI algorithms evaluate these and identify areas for improvement. It receives outlier data as input and provides the results of wasteful cost identification as output. For example, it identifies unnecessarily high communication costs and analyzes the causes.

[1084] Step 6:

[1085] cost-cutting measures

[1086] The server generates specific cost-saving proposals based on the waste identification results. It also simulates the effectiveness of the proposals in advance and sets priorities. It uses the waste identification results and related data as input and provides cost-saving proposals and their predicted effects as output. For example, it proposes specific measures for optimizing energy use and prioritizes the proposals based on their importance.

[1087] Step 7:

[1088] Implementation Plan

[1089] The server then develops a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments. It uses the cost-saving proposals as input and provides a detailed implementation plan as output. For example, an energy reduction plan can be developed, which the user can review and adjust to fit their office hours.

[1090] Step 8:

[1091] Cost reduction implementation

[1092] The device performs the execution activities based on the formulated plan. The server monitors this in real time and records the progress. It uses the implementation plan as input and generates execution progress data as output. For example, a user implements new energy usage guidelines and the server monitors their progress.

[1093] Step 9:

[1094] Confirmation of effectiveness

[1095] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements. If necessary, it will propose further optimizations. It uses the execution result data as input and provides the results of the effectiveness check and further improvement suggestions as output. For example, it can confirm the effectiveness of a new energy usage method and propose further improvements.

[1096] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1097] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management. In particular, it is characterized by combining an emotion engine that recognizes the user's emotions and providing customized suggestions according to the user's emotions.

[1098] The system of the present invention includes the following elements:

[1099] 1. Data Collection

[1100] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data.

[1101] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[1102] 2. Data cleansing

[1103] The server checks the consistency and completeness of the retrieved data, imputes missing values ​​using predictive algorithms, and standardizes the format.

[1104] Example: The server corrects mismatched date formats and missing values.

[1105] 3. Cost Analysis

[1106] The server performs statistical processing to identify outliers in expenses and then performs trend analysis to predict future costs.

[1107] Example: A server analyzes energy consumption data and predicts consumption trends for the next year.

[1108] 4. Categorization

[1109] The server automatically categorizes expense data and allows users to review and adjust the details.

[1110] Example: A server categorizes expenses by department or project.

[1111] 5. Identify waste

[1112] The server uses an anomaly detection algorithm to quickly identify wasteful costs, and AI evaluates the potential for improvement.

[1113] Example: A server identifies high communication charges and analyzes their causes.

[1114] 6. Cost reduction proposals

[1115] The server generates specific cost-cutting proposals based on the identified waste, simulates the effectiveness of the proposals, and sets priorities.

[1116] Example: Propose specific measures to improve the energy efficiency of a server and list them in order of importance.

[1117] Additionally, the system of the present invention incorporating the emotion engine includes the following elements:

[1118] 1. Emotion recognition

[1119] The server uses an emotion engine to recognize the user's emotions and analyzes the user's writings and comments to identify the user's emotional state.

[1120] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[1121] 2. Emotion-based personalized recommendations

[1122] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. For example, if the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[1123] Example: When a user is feeling stressed, the server prioritizes suggestions for reducing stress that are relatively easy to implement.

[1124] 3. Emotional Feedback

[1125] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[1126] Example: The server displays a dashboard showing the user's current emotional state and provides suggestions for improvement based on that.

[1127] For example, if a company's human resources manager is feeling stressed, the system will recognize that emotion and first offer relatively simple and effective improvement suggestions, allowing the manager to start with suggestions that are easy to implement. The system also monitors the effectiveness of the suggestions after implementation and, if further suggestions are needed, suggests next steps based on the manager's emotional state.

[1128] In this way, the present invention aims to make corporate cost management more effective and user-friendly by combining AI technology and emotion recognition technology.

[1129] The processing flow will be explained below.

[1130] Step 1:

[1131] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data, which includes verifying credentials and specifying the time period to retrieve.

[1132] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[1133] Step 2:

[1134] The server checks the integrity and completeness of the retrieved data, adapts to the data format and content, identifies missing or outlier values, and imputes data as needed.

[1135] Example: Standardizing the date format and number of digits in data collected by the server, and detecting and correcting inappropriate values.

[1136] Step 3:

[1137] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[1138] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[1139] Step 4:

[1140] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[1141] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[1142] Step 5:

[1143] The server automatically categorizes expense data by category, allows users to perform detailed analysis for each specific category, and allows users to manually adjust the categorization results as needed.

[1144] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[1145] Step 6:

[1146] The server uses AI technology to identify outliers, using predictive algorithms to spot spending items that fall outside of the normal range.

[1147] Example: A server detects extremely high communication costs and analyzes the background.

[1148] Step 7:

[1149] The server generates specific cost-saving proposals based on the identified outliers, simulates the effects of the proposals, and provides the proposals in an easy-to-implement format for the user.

[1150] Example: Propose specific measures to improve the energy efficiency of servers.

[1151] Step 8:

[1152] The server prioritizes the generated reduction proposals, allowing the user to implement the most effective proposals first.

[1153] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[1154] Step 9:

[1155] The server creates a detailed implementation plan based on the reduction proposals, including specific steps and schedules, which the user can review and adjust.

[1156] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[1157] Step 10:

[1158] The terminal executes cost-cutting activities based on the established plan, and the server monitors the progress in real time and provides feedback to the user.

[1159] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[1160] Step 11:

[1161] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[1162] Example: A server checks the effectiveness of new energy usage methods and suggests further improvements.

[1163] (A system incorporating an emotion engine)

[1164] Step 12:

[1165] The server uses an emotion engine to recognize the user's emotions by analyzing the user's posts and comments to identify their emotional state.

[1166] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[1167] Step 13:

[1168] The server customizes the content and priorities of cost-saving proposals based on the user's recognized emotions, giving priority to proposals that are less burdensome, taking into account the user's emotional state.

[1169] Example: When a server detects a user is feeling stressed, it prioritizes suggestions for reducing stress that are relatively easy to implement.

[1170] Step 14:

[1171] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[1172] Example: The server displays a dashboard showing the user's current emotional state and provides appropriate improvement suggestions based on that.

[1173] In this way, a system that combines an emotion engine can make flexible cost optimization proposals that take into account the user's emotional state, aiming to make corporate cost management more efficient and user-friendly.

[1174] Example 2

[1175] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1176] Conventional cost management systems collect and analyze expense data, generate cost-reduction proposals, and formulate implementation plans, but they are unable to provide effective proposals that take into account the user's emotional state. This can lead to stress and difficulty in implementing the proposals. Therefore, there is a need for a system that recognizes the user's emotional state and provides customized proposals based on that state.

[1177] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1178] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost-cutting proposals based on the identified outliers, means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness, means for identifying the user's emotional state using an emotion engine that recognizes the user's emotions, means for customizing the content and priorities of the cost-cutting proposals based on the user's emotional state, and means for visually providing feedback on the user's emotional state, thereby enabling effective and less burdensome cost-cutting proposals to be made in accordance with the user's emotional state.

[1179] "Data sources" refers to the systems and applications that store a company's expense data, including ERP systems and accounting software.

[1180] "Expense data" is data that contains detailed information about the costs a company incurs on a day-to-day basis.

[1181] "Integrity" means that data is accurate, consistent, and does not contradict each other.

[1182] "Complete" means that the data contains all necessary information and is free of missing parts.

[1183] "Imputation" refers to filling in missing information in data using prediction or imputation algorithms.

[1184] An "outlier" is a value that is extremely outlier compared to other values ​​in a data set, and often includes errors or exceptions.

[1185] A "cost reduction proposal" refers to a proposal that outlines specific action items and strategies for reducing a company's expenses.

[1186] "Plan" refers to a document that contains specific steps and timelines for implementing the generated cost reduction proposals.

[1187] "Progress" refers to the status of the implementation of planned specific actions, and indicates how much progress has been made over time.

[1188] "Effects" refer to the results obtained as a result of implementing a plan, and specifically include the degree of cost reduction and improved efficiency.

[1189] "Emotion engine" refers to software equipped with artificial intelligence technology to recognize and analyze a user's emotional state.

[1190] "Emotional state" refers to the psychological state such as stress or satisfaction felt by the user.

[1191] "Feedback" refers to providing the analysis results and suggestions to the user visually or in other ways, allowing the user to check their own status and the suggestions.

[1192] This invention is a system that utilizes AI technology to improve the efficiency of corporate cost management, and is characterized by its incorporation of an emotion engine that recognizes user emotions. To implement this system, hardware and software including the following elements are used.

[1193] 1. Data Collection

[1194] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data. Specifically, the server sends an HTTP request and receives data in JSON format from the ERP system or accounting software.

[1195] Example: A server sends a request to "https: / / api.erp.example.com / v1 / expenses?month=last" to retrieve last month's expense data from a company's ERP system.

[1196] 2. Data cleansing

[1197] The server checks the integrity and completeness of the data retrieved, imputes missing values ​​using predictive algorithms, and standardizes formats, including detecting duplicate entries in the database and using a reverse dictionary to infer missing dates.

[1198] Example: A server corrects inconsistencies in date formats between expense data entries and imputes missing date fields with a predictive algorithm.

[1199] 3. Cost Analysis

[1200] The server performs statistical processing to identify outliers in expenses and forecast future costs. Specifically, it uses the Pandas library to detect outliers and the Prophet model to forecast future cost trends.

[1201] Example: A server analyzes energy consumption data and predicts energy consumption trends for the next year.

[1202] 4. Categorization

[1203] The server automatically categorizes expense data into categories and allows users to review and adjust the details, including using machine learning models to categorize data into specific categories.

[1204] Example: A server organizes expense data into categories such as communication expenses, travel expenses, and advertising expenses.

[1205] 5. Identify waste

[1206] The server uses anomaly detection algorithms to identify wasteful costs and evaluate potential areas for improvement, using techniques such as Isolation Forest and One-Class SVM.

[1207] Example: The server identifies communication expenses that are significantly higher than average as an anomaly and analyzes the details.

[1208] 6. Cost reduction proposals

[1209] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[1210] Example: A server generates a list of proposals for replacing lighting with LEDs to improve energy efficiency, based on estimated results.

[1211] 7. Emotion recognition

[1212] The server uses an emotion engine to recognize user emotions and identify the emotional state from user posts and statements, using a natural language processing model (e.g., BERT).

[1213] Example: A server analyzes a user's message "Work has been very stressful lately" and determines that the user is feeling stressed.

[1214] 8. Emotion-based personalized recommendations

[1215] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. If the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[1216] Example: When a user feels stressed, the server displays a low-impact reduction suggestion such as "reduce the frequency of meetings."

[1217] 9. Emotional Feedback

[1218] The server provides visual feedback of the user's emotional state, allowing the user to check the state and select suggestions using a visualization tool.

[1219] Example: The server displays the user's emotional state on a dashboard as "Stress level: High" and provides suggestions for improvement based on that.

[1220] This system allows companies to achieve efficient and user-friendly cost management, allowing users to receive cost-saving suggestions that take their emotional state into account and are easy to implement.

[1221] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1222] Program processing flow

[1223] Step 1: Data collection

[1224] The server contacts the data source and sends an API request to retrieve expense data.

[1225] Input: API endpoint URL and authentication information configured on the server.

[1226] Specific behavior: The server sends an HTTP request to an endpoint such as "https: / / api.erp.example.com / v1 / expenses?month=last".

[1227] Output: Expense data in JSON format returned from your ERP system or accounting software.

[1228] Step 2: Data cleansing

[1229] The server checks the consistency and completeness of the data it retrieves, fills in missing values, and standardizes the format.

[1230] Input: Captured expense data in JSON format.

[1231] What it does: The server uses a database library to read data, detects duplicate entries, fills in missing data with a predictive algorithm, and standardizes data formats (for example, standardizing date formats to "YYYY-MM-DD").

[1232] Output: A clean dataset that has been checked for consistency and completeness, imputed for missing values, and has a uniform format.

[1233] Step 3: Cost analysis

[1234] The server analyzes the cleansed data to identify outliers and predict future costs.

[1235] Input: The cleansed dataset.

[1236] What it does: The server uses a statistical library to detect outliers in the data and a time series forecasting model (e.g., Prophet) to predict future cost trends.

[1237] Output: Identified outliers and future cost forecast data.

[1238] Step 4: Categorization

[1239] The server automatically categorizes expense data and allows users to review and adjust the details.

[1240] Input: The cleansed dataset.

[1241] What it does: The server uses machine learning models to categorize expense data by category (e.g., communication, travel, advertising).

[1242] Output: Expense data broken down by category.

[1243] Step 5: Identify waste

[1244] The server uses an anomaly detection algorithm to identify wasteful costs and evaluate potential for improvement.

[1245] Input: Expense data broken down by category.

[1246] How it works: The server uses outlier detection algorithms such as Isolation Forest and One-Class SVM to identify unreasonably high expenditures.

[1247] Output: Identified waste costs and a detailed report.

[1248] Step 6: Cost reduction proposal

[1249] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[1250] Input: Detailed data on waste costs.

[1251] Specific operation: The server generates improvement measures using heuristic rules and verifies their effectiveness using a simulation model.

[1252] Output: Prioritized cost-saving proposals and their predicted effectiveness.

[1253] Step 7: Emotion Recognition

[1254] The server uses an emotion engine to recognize the user's emotions and analyzes the user's emotional state from their statements and writings.

[1255] Input: User message and feedback data.

[1256] What happens: The server analyzes the text data using a natural language processing model (e.g., BERT) to identify the user's emotional state.

[1257] Output: User's emotional state report.

[1258] Step 8: Customize your suggestions based on emotions

[1259] The server customizes the suggestions and priorities based on the user's emotions recognized.

[1260] Input: User emotional state report, cost reduction suggestions.

[1261] Specific operation: The server selects and customizes suggestions that are easy to implement and have high priority based on the user's emotional state.

[1262] Output: Customized cost-saving proposals.

[1263] Step 9: Emotional Feedback

[1264] The server visually feeds back the user's emotional state, allowing the user to confirm the state and select suggestions.

[1265] Input: customized cost-saving suggestions, user emotional state report.

[1266] Specific operation: The server uses visualization tools such as a dashboard to display the user's emotional state and suggestions.

[1267] Output: Emotional feedback and improvement suggestions on a visualized dashboard.

[1268] (Application example 2)

[1269] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1270] Conventional cost management systems collect and analyze expense data, identify outliers, and generate cost-reduction proposals. However, these processes do not take into account the user's emotional state, often resulting in stress. Furthermore, in logistics centers, worker comfort and efficiency are important, and workload adjustments based on individual emotional states are required. However, current systems lack an effective means of doing this, resulting in insufficient improvements in work efficiency and reductions in worker stress.

[1271] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost reduction proposals based on the identified outliers, means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness, means for determining the user's emotional state using emotion recognition technology and customizing the content and priority of the proposals based on the determined emotional state, and means having a human interface for presenting the customized proposals to the user. This enables efficient cost management that takes the user's emotional state into consideration, reducing worker stress and improving work efficiency at logistics centers.

[1272] "Data sources" refers to information sources such as ERP systems or accounting software that store a company's expense data.

[1273] "Expense data" is detailed information about the costs a company incurs on a daily basis, including travel, communication, and energy costs.

[1274] "Integrity" refers to data that is consistently accurate and consistent across different data sets.

[1275] "Complete" refers to the state in which all data is available and there are no missing values ​​or incomplete information.

[1276] "Imputation" refers to the process of estimating and adding missing data to complete a dataset.

[1277] An "outlier" is a data value that deviates from normal patterns or acceptable ranges.

[1278] "Cost reduction proposals" refer to specific action plans and measures to reduce a company's expenses.

[1279] A "plan" is a schedule that defines the steps and resources needed to achieve a specific goal.

[1280] "Progress" refers to the status of execution and degree of success of a plan or project.

[1281] "Effect" refers to the results or outcomes that result from a certain action or measure.

[1282] "Emotion recognition technology" refers to technology that identifies a user's emotional state (e.g., stress, joy, anger, etc.) from facial expressions and vocal tone.

[1283] "User" refers to the person who operates the system and manages and analyzes expense data.

[1284] "Emotional state" refers to a user's current psychological state as determined by emotion recognition technology.

[1285] "Human interface" refers to the interface through which a user interacts with a system and exchanges information.

[1286] This invention provides a system aimed at efficient cost management and worker stress reduction in logistics centers. This system utilizes AI technology and emotion recognition technology to make suggestions based on the user's emotional state. Specific embodiments are described below.

[1287] The system consists of a server and smart glasses as its main components. The server automatically collects expense data from data sources such as a company's ERP system or accounting software, including the means to retrieve the required data using API requests.

[1288] The server checks the consistency and completeness of collected expense data, fills in missing values ​​with a predictive algorithm, and standardizes the format, thereby ensuring data quality.

[1289] The server then performs statistical processing and analyses on the collected and stored data, identifying outliers and conducting trend analysis to predict future costs, which in turn generates cost-saving recommendations.

[1290] The server uses emotion recognition technology to collect data through the smart glasses to determine the worker's emotional state. Specifically, it uses facial recognition and voice analysis to determine the worker's emotional state from their facial expressions and tone. Based on the determined emotional state, the server customizes the content and priorities of suggestions.

[1291] The customized suggestions are sent from the server to the smart glasses and presented to the worker through a human interface, so that suggestions for a lighter workload are displayed preferentially to workers who are feeling stressed.

[1292] The server uses TensorFlow and OpenCV to build emotion recognition models, Flask framework to manage data on the server side, and SQLite to operate a local database. Communication with the smart glasses is via Bluetooth Low Energy (BLE).

[1293] For example, if a worker at a logistics center is feeling stressed, the system will recognize that emotion and suggest lighter tasks first, such as avoiding carrying heavy items and prioritizing stocking lighter items, thereby reducing the worker's stress and allowing them to work efficiently.

[1294] Below is an example of a prompt sentence that uses a generative AI model to design suggestions based on the worker's emotions.

[1295] Example prompt sentence:

[1296] Consider a graphical interface design that uses an emotion engine to identify the emotions of workers and suggest lighter tasks to stressed workers. In particular, be sure to consider the use of colors and layout to reduce stress.

[1297] In this way, this system simultaneously improves the cost performance and work efficiency of logistics centers through efficient management of expense data and suggestions based on the emotional state of workers.

[1298] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1299] Step 1:

[1300] The server collects expense data from data sources, specifically, from ERP systems and accounting software using API requests. The input is the access information of the data source based on the API request, and the output is the collected raw expense data.

[1301] Step 2:

[1302] The server checks the collected expense data for consistency and completeness and imputes it if necessary, correcting inconsistent date formats and imputing missing values ​​using predictive algorithms. The input is the raw expense data collected, and the output is cleaned data that has been checked for consistency and completeness.

[1303] Step 3:

[1304] The server analyzes the collected and imputed data to identify outliers. It performs statistical processing to detect outliers that deviate from the normal range. The input is the cleaned expense data, and the output is a list of outliers.

[1305] Step 4:

[1306] The server generates cost reduction proposals based on the identified outliers. It analyzes the causes of the outliers and proposes specific cost reduction measures accordingly. The input is a list of outliers, and the output is the generated cost reduction proposals.

[1307] Step 5:

[1308] The server uses emotion recognition technology to determine the worker's emotional state. It uses the camera and microphone in the smart glasses to perform facial recognition and voice analysis to identify the worker's emotional state. The input is real-time video and audio data, and the output is the determined emotional state.

[1309] Step 6:

[1310] The server customizes the content and priority of the proposals based on the determined emotional state. If the user is feeling stressed, it prioritizes proposals that are less stressful. The input is the determined emotional state and the generated cost-saving proposals, and the output is the customized proposals.

[1311] Step 7:

[1312] The server sends the customized suggestions to the smart glasses for presentation to the worker. The suggestions are displayed to the worker via a human interface using Bluetooth Low Energy (BLE). The input is the customized suggestions, and the output is the suggestions displayed on the smart glasses display.

[1313] Step 8:

[1314] The server develops an implementation plan for the proposals and monitors their progress and effectiveness. It evaluates the results of implementing the proposed cost-saving measures and obtains feedback. The inputs are the implemented proposals and their results data, and the outputs are progress reports and effectiveness evaluation reports.

[1315] By carrying out the above processing steps, efficient cost management and stress reduction for workers at the logistics center can be achieved.

[1316] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1317] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1318] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1319] [Fourth embodiment]

[1320] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1321] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1322] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1323] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1324] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1325] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1326] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1327] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1328] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1329] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1330] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1331] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1332] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1333] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[1334] The system of the present invention achieves cost optimization mainly through the following steps.

[1335] 1. Data Collection

[1336] The server automatically collects the company's expense data from data sources such as ERP systems, accounting software, etc. It sends API requests to retrieve the required data.

[1337] Example: A server accesses a company's ERP system and downloads last month's expense report.

[1338] 2. Data cleansing

[1339] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes formats, and corrects invalid values.

[1340] Example: Standardizing the date format of data collected by the server and correcting incorrect values.

[1341] 3. Cost Analysis

[1342] The server performs basic statistical processing on the organized data to identify outliers, and also performs trend analysis on past data to predict future costs.

[1343] Example: A server analyzes energy consumption data from the past 12 months and predicts consumption trends for the next year.

[1344] 4. Categorization

[1345] The server automatically categorizes expense data and allows users to analyze it by specific categories, with the option to manually adjust the categorization as needed.

[1346] Example: The server categorizes collected expense data by department and project, and the user adjusts it appropriately.

[1347] 5. Identify waste

[1348] The server quickly identifies wasteful costs and excessive spending based on outliers, which are then evaluated by AI algorithms to highlight areas for improvement.

[1349] Example: A server identifies unnecessarily high communication costs and analyzes the cause.

[1350] 6. Cost-cutting measures

[1351] The server generates specific cost-cutting proposals based on the results of waste identification, and also simulates the effects of the proposals in advance to set priorities.

[1352] Example: The server proposes specific measures for optimizing energy usage and prioritizes them based on their importance.

[1353] 7. Implementation Plan

[1354] The server then creates a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments.

[1355] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[1356] 8. Cost reduction implementation

[1357] The terminal carries out the execution activities based on the formulated plan, and the server monitors this in real time and records the progress.

[1358] Example: A user implements new energy usage guidelines and a server monitors their progress.

[1359] 9. Confirmation of effectiveness

[1360] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[1361] Example: A server checks the effectiveness of new energy usage methods and suggests further improvements.

[1362] Each step of this system automates a company's cost management and improves operational efficiency by accurately and quickly identifying and eliminating waste.

[1363] The processing flow will be explained below.

[1364] Step 1:

[1365] The server accesses a data source, such as an ERP system or accounting software, and sends an API request to retrieve expense data, which includes verifying credentials and specifying the time period to retrieve.

[1366] Example: A server accesses a company's ERP system and downloads last month's expense data using an API.

[1367] Step 2:

[1368] The server checks the integrity and completeness of the retrieved data, adapting to the data format and content and identifying missing or outlier values.

[1369] Example: Standardizing the date format and number of digits in data collected by the server and detecting inappropriate values.

[1370] Step 3:

[1371] The server uses statistical models and predictive algorithms to impute missing values, and also standardizes and cleans the data.

[1372] Example: A server uses a predictive model to fill in missing energy usage data and correct improper formatting.

[1373] Step 4:

[1374] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[1375] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[1376] Step 5:

[1377] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[1378] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[1379] Step 6:

[1380] The server automatically categorizes expense data by category, allowing users to perform detailed analysis by specific category and sorting the data.

[1381] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[1382] Step 7:

[1383] The server identifies wasteful costs using an outlier detection algorithm, quickly uncovering expenditure items that fall outside of normal ranges.

[1384] Example: A server detects extremely high communication costs and analyzes the background.

[1385] Step 8:

[1386] The server generates specific cost-cutting proposals based on the identified waste, and the AI ​​automatically devise effective improvement methods and provides them to the user.

[1387] Example: Propose specific measures to improve the energy efficiency of servers.

[1388] Step 9:

[1389] The server assigns a priority to each suggestion, allowing the user to execute the most effective suggestion first.

[1390] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[1391] Step 10:

[1392] The server creates an action plan based on the reduction proposals, generates a plan including detailed procedures and schedules, and presents it to the user.

[1393] Example: The server creates a specific action plan for reducing energy usage, which the user confirms.

[1394] Step 11:

[1395] The terminal executes cost-cutting activities based on the established plan, and the server monitors this process in real time and records the progress.

[1396] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[1397] Step 12:

[1398] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[1399] Example: A server checks the effectiveness of new energy usage methods and suggests further improvements.

[1400] As described above, by performing specific operations at each step, a system is provided that realizes cost optimization for a company.

[1401] Example 1

[1402] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1403] Current corporate cost management systems often involve manual data collection and analysis processes, resulting in a lack of efficiency. It also makes it difficult to detect wasteful costs and outliers early on, making it difficult to propose effective cost-cutting measures. Furthermore, the consistency and completeness of the collected data cannot be guaranteed, making it difficult to obtain reliable analysis results. To solve these problems and improve corporate operational efficiency, automated data collection, cleansing, and analysis processes are needed.

[1404] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1405] In this invention, the server includes means for automatically collecting expense data from data sources, means for verifying the consistency and completeness of the collected expense data and imputing missing values ​​using a predictive model, means for performing statistical processing on the collected and imputed data to identify outliers, means for generating cost reduction proposals using an AI algorithm based on the identified outliers, means for simulating the effects of the generated proposals in advance and setting priorities, and means for formulating action plans based on the priorities and monitoring their progress and effectiveness in real time, thereby enabling companies to efficiently collect and analyze data and quickly identify and reduce wasteful costs.

[1406] "Data source" refers to the systems and software that provide a company's expense data.

[1407] "Expense data" refers to data that includes information about expenses incurred by a company in various activities.

[1408] "Integrity" refers to a state in which data is consistent and free of contradictions.

[1409] "Complete" refers to the state in which data contains all necessary information.

[1410] A "predictive model" refers to an algorithm or mathematical model that estimates future values ​​based on trends and patterns in data.

[1411] "Statistical processing" refers to methods for summarizing and characterizing data as part of data analysis.

[1412] An "outlier" is a value in the data that is significantly different from the other values.

[1413] "AI algorithm" refers to a data processing method based on machine learning and artificial intelligence.

[1414] "Cost reduction proposals" refer to specific methods and measures for reducing unnecessary costs.

[1415] "Simulation" refers to the process of testing the effectiveness of a proposal in a virtual environment.

[1416] "Priority" refers to ranking multiple items or proposals according to their importance.

[1417] "Implementation plan" refers to a plan for specifically implementing the proposed cost reduction measures.

[1418] "Real-time" refers to a state in which processing and data collection occur immediately.

[1419] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management and reduce unnecessary costs. The system provides a consistent process from data collection to the implementation and monitoring of cost reduction plans.

[1420] The main hardware of the system is a server, which automatically collects expense data from data sources such as a company's ERP system and accounting software. Specifically, it retrieves the necessary data from these sources by sending API requests. The server also checks the consistency and completeness of the collected data and imputes missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The server also standardizes the data format and corrects inappropriate values.

[1421] The server then performs statistical operations on the organized data to identify outliers. For example, it uses Pandas to calculate the standard deviation of the data frame to identify outliers. The server also performs trend analysis on the historical data and uses forecasting models such as ARIMA models to forecast future costs.

[1422] The server then automatically categorizes expense data by department and project, allowing users to manually adjust this as needed. It uses Python regular expressions to parse expense record descriptions and classify them into the appropriate categories. Users can review and modify the categorization results using a web interface.

[1423] To identify wasteful costs, the server uses AI algorithms (e.g., random forests) to assess wasteful costs and excessive spending based on outliers, allowing for quick identification of areas for improvement.

[1424] The server also generates specific cost-reduction proposals based on the results of identifying waste. The generated proposals have the ability to simulate their effects in advance and set priorities. For example, the server generates a proposal for installing new lighting equipment and simulates its energy-saving effects. It also calculates the expected cost savings for each proposal and creates a priority list.

[1425] Furthermore, the server creates an action plan based on the priority, which the user can review and ultimately adjust.Specific action plans for each department are created in Excel files, which the user can review and modify via a web interface.

[1426] Finally, the device carries out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. For example, a user implements new energy usage guidelines, and the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard. After this implementation, the server re-analyzes the post-improvement data to confirm the effectiveness of the implementation of the new energy usage guidelines.

[1427] Example prompts to input to a generative AI model:

[1428] "Please explain a program that collects expense data from an ERP system, cleanses the data by imputing missing values ​​with a predictive model, identifies outliers through statistical processing and trend analysis, generates optimization proposals, and formulates specific implementation plans."

[1429] Through this system, companies can streamline cost management and reduce wasteful spending.

[1430] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1431] Step 1:

[1432] Data collection

[1433] The server automatically collects expense data from ERP systems and accounting software. Specifically, the server sends API requests to retrieve the required data from these data sources. The input is the API endpoint of each data source, and the output is the collected expense data. For example, the server sends a request such as "GET / api / expenses?month=2023-09" to the ERP system, receives expense data as a response, and stores it in the database.

[1434] Step 2:

[1435] Data Cleansing

[1436] The server checks the consistency and completeness of the collected expense data and imputes missing values. Specifically, it predicts missing values ​​using a predictive model (e.g., Scikit-learn's SimpleImputer). The input is the collected raw expense data, and the output is cleansed data with missing values ​​imputed and a unified format. The server runs a script to unify the date format to "YYYY-MM-DD" and corrects inappropriate values.

[1437] Step 3:

[1438] Cost Analysis

[1439] The server performs statistical processing on the cleansed data to identify outliers. The input is the cleansed expense data, and the output is an analysis showing outliers and trends. For example, the server uses Pandas to calculate the standard deviation of the data frame to identify outliers. It also uses an ARIMA model to forecast future cost trends using data from the past 12 months.

[1440] Step 4:

[1441] Category Classification

[1442] The server categorizes expense data by department or project, allowing users to manually adjust as needed. The input is the analyzed expense data, and the output is the categorized expense data. The server uses Python regular expressions to parse the expense record description and classify it into the appropriate category. Users can view the categorization results through a web interface and make manual adjustments.

[1443] Step 5:

[1444] Identifying waste

[1445] The server identifies wasteful costs and excessive expenditures based on outliers. The input is the analysis results, including outliers, and the output is a list of wasteful costs. Specifically, the server uses an AI algorithm (e.g., random forest) to evaluate wasteful expenditures. The server detects abnormally high communication costs and displays them in a histogram.

[1446] Step 6:

[1447] cost-cutting measures

[1448] The server generates cost-saving proposals based on the results of waste identification. The input is a list of wasteful costs, and the output is specific cost-saving proposals. The server generates proposals for installing new lighting equipment and simulates their energy-saving effects. It calculates the expected cost savings for each proposal and creates a priority list.

[1449] Step 7:

[1450] Implementation Plan

[1451] The server creates an action plan for cost-cutting proposals, which the user can review and adjust. The input is the cost-cutting proposals, and the output is a specific implementation plan. For example, the server creates a specific action plan for each department in an Excel file, and the user can review the plan via a web interface and modify it as needed.

[1452] Step 8:

[1453] Cost reduction implementation

[1454] The terminals carry out the implementation activities based on the formulated plan, and the server monitors the progress and effects in real time. The input is the implementation plan, and the output is progress data of the implementation activities. For example, when a user puts new energy usage guidelines into practice, the server collects real-time energy consumption data through IoT devices and displays the progress on a dashboard.

[1455] Step 9:

[1456] Confirmation of effectiveness

[1457] The server re-analyzes the post-implementation data and confirms the effects of the improvements. The input is the post-implementation data, and the output is the analysis of the effects. The server analyzes the data after the new energy usage guidelines are implemented and displays the effects in a graph. The server then presents the user with a report showing the potential for further cost reductions.

[1458] Through this series of steps, companies can streamline cost management and quickly identify and reduce wasteful spending.

[1459] (Application example 1)

[1460] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1461] Corporate cost management is complex, time-consuming, and labor-intensive, often resulting in a lot of unnecessary expenses. Furthermore, logistics centers often lack real-time information on resource usage, making efficient cost reduction difficult. The present invention aims to solve these problems by providing a system for efficiently managing corporate expenses and reducing unnecessary costs.

[1462] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1463] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost reduction proposals based on the identified outliers, means for formulating plans for implementing the generated proposals and monitoring their progress and effectiveness, and means for understanding the resource usage status of the logistics center in real time via a smartphone application and quickly identifying and reducing unnecessary costs. This automates corporate expense management and enables real-time cost management and reduction at logistics centers.

[1464] A "data source" is an information system or database that provides a company's expense data.

[1465] "Expense data" refers to data that includes all expense information related to the day-to-day running of a company.

[1466] "Integrity" refers to a state in which data is consistent and free of contradictions.

[1467] "Complete" refers to the state in which all data is present and without any missing data.

[1468] "Completion" refers to the process of filling in missing data using predictive models, etc.

[1469] An "outlier" is a value that is significantly different from the rest of the data and is outside the range of what is normally expected.

[1470] A "cost reduction proposal" is a specific measure or action plan that should be implemented to reduce wasteful expenses.

[1471] "Plan" refers to the design of specific steps and actions to implement a proposal.

[1472] "Progress" is a state or condition that indicates how far a planned action has been carried out.

[1473] "Effectiveness" is the result that shows how much cost savings the implemented proposals and plans actually achieved.

[1474] A "smartphone application" is a software program that runs on a smartphone and acts as a user interface to provide real-time information and enable operation.

[1475] A "logistics center" is a base for storing, managing, and delivering goods, and is a facility that carries out efficient logistics activities.

[1476] "Resource usage" refers to the utilization of energy, labor, equipment, etc. within a logistics center.

[1477] This invention relates to a system for efficiently managing corporate costs and resource usage at a logistics center. This system can be realized using a smartphone application and a server.

[1478] 1. System Configuration

[1479] The server automatically collects expense data from data sources and checks the consistency and completeness of the collected expense data. It supplements missing data as needed and analyzes the collected data to identify outliers. It generates cost-cutting proposals based on the outliers, develops detailed plans for implementing the proposals, and monitors their progress and effectiveness. Furthermore, a smartphone application allows users to grasp the distribution center's resource usage in real time, quickly identifying and reducing unnecessary costs.

[1480] 2. Hardware and Software

[1481] The server uses the following hardware and software:

[1482] Hardware: A server with a powerful processor and ample storage capacity

[1483] Software: ERP system and accounting software APIs for data collection, pandas and NumPy for data cleansing, TensorFlow and Scikit-learn for statistical processing and outlier detection, Flask / Django as web application frameworks, Firebase / Firestore as real-time databases

[1484] The smartphone application has the following features:

[1485] Data collection: Collect data from ERP systems, sensors in distribution centers, and barcode readers.

[1486] Real-time monitoring: View resource usage in your distribution center in real time

[1487] Notification function: Alerts when unnecessary costs or abnormal values ​​occur

[1488] 3. Program Processing Overview

[1489] The server cleanses expense data collected from data sources for consistency and completeness, and uses predictive models to fill in any missing data. It then performs basic statistical processing on the cleansed data to identify outliers. Based on the identified outliers, an AI algorithm generates cost-saving proposals and performs simulations to verify their effectiveness. Finally, it develops prioritized cost-saving plans, and monitors their progress and effectiveness in real time.

[1490] 4. Examples and prompts

[1491] For example, if you want to predict consumption trends for the next year based on a logistics center's energy consumption data from the past 12 months, you can input the following prompt into the generative AI model:

[1492] "Given the last 12 months of energy consumption data for a distribution center, please generate a model that will predict consumption trends for the next fiscal year. Also, please generate and prioritize specific proposals for cost reduction."

[1493] Using this prompt, the AI ​​model generates output through the following steps:

[1494] 1. Preprocessing of input data

[1495] 2. Trend Forecasting

[1496] 3. Outlier detection

[1497] 4. Cost reduction proposals

[1498] 5. Prioritization

[1499] This system will significantly improve the efficiency of cost management at logistics centers and enable real-time cost reductions.

[1500] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1501] Step 1:

[1502] Data collection

[1503] The server collects expense and resource usage data from the ERP system, sensors in the distribution center, and barcode readers. As input, it sends API requests to retrieve the required data. For example, it accesses the ERP system to download expense reports for the past 12 months and retrieves real-time energy consumption data from sensors. As output, these data are passed to subsequent processing steps.

[1504] Step 2:

[1505] Data Cleansing

[1506] The server checks the consistency and completeness of the collected data, imputes missing values ​​using predictive models, standardizes the date format of the data, and corrects improper values. It receives the collected raw data as input and produces cleansed and organized data as output.

[1507] Step 3:

[1508] Cost Analysis

[1509] The server performs basic statistical processing on the cleaned data to identify outliers. It also performs trend analysis on past data to predict future costs. It takes the cleansed data as input and provides outliers and predictions of future cost trends as output. For example, it analyzes energy consumption data from the past 12 months to predict consumption trends for the next year.

[1510] Step 4:

[1511] Category Classification

[1512] The server automatically categorizes expense data and allows users to analyze it by specific categories. Users can also manually adjust the categorization results as needed. It uses the analyzed data as input and generates categorized data as output. For example, categorizing expense data by department or project.

[1513] Step 5:

[1514] Identifying waste

[1515] The server quickly identifies wasteful costs and excessive expenditures based on outliers. AI algorithms evaluate these and identify areas for improvement. It receives outlier data as input and provides the results of wasteful cost identification as output. For example, it identifies unnecessarily high communication costs and analyzes the causes.

[1516] Step 6:

[1517] cost-cutting measures

[1518] The server generates specific cost-saving proposals based on the waste identification results. It also simulates the effectiveness of the proposals in advance and sets priorities. It uses the waste identification results and related data as input and provides cost-saving proposals and their predicted effects as output. For example, it proposes specific measures for optimizing energy use and prioritizes the proposals based on their importance.

[1519] Step 7:

[1520] Implementation Plan

[1521] The server then develops a detailed implementation plan for implementing the generated proposals, which the user can use to make final adjustments. It uses the cost-saving proposals as input and provides a detailed implementation plan as output. For example, an energy reduction plan can be developed, which the user can review and adjust to fit their office hours.

[1522] Step 8:

[1523] Cost reduction implementation

[1524] The device performs the execution activities based on the formulated plan. The server monitors this in real time and records the progress. It uses the implementation plan as input and generates execution progress data as output. For example, a user implements new energy usage guidelines and the server monitors their progress.

[1525] Step 9:

[1526] Confirmation of effectiveness

[1527] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements. If necessary, it will propose further optimizations. It uses the execution result data as input and provides the results of the effectiveness check and further improvement suggestions as output. For example, it can confirm the effectiveness of a new energy usage method and propose further improvements.

[1528] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1529] This invention relates to a system that utilizes AI technology to improve the efficiency of corporate cost management. In particular, it is characterized by combining an emotion engine that recognizes the user's emotions and providing customized suggestions according to the user's emotions.

[1530] The system of the present invention includes the following elements:

[1531] 1. Data Collection

[1532] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data.

[1533] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[1534] 2. Data cleansing

[1535] The server checks the consistency and completeness of the retrieved data, imputes missing values ​​using predictive algorithms, and standardizes the format.

[1536] Example: The server corrects mismatched date formats and missing values.

[1537] 3. Cost Analysis

[1538] The server performs statistical processing to identify outliers in expenses and then performs trend analysis to predict future costs.

[1539] Example: A server analyzes energy consumption data and predicts consumption trends for the next year.

[1540] 4. Categorization

[1541] The server automatically categorizes expense data and allows users to review and adjust the details.

[1542] Example: A server categorizes expenses by department or project.

[1543] 5. Identify waste

[1544] The server uses an anomaly detection algorithm to quickly identify wasteful costs, and AI evaluates the potential for improvement.

[1545] Example: A server identifies high communication charges and analyzes their causes.

[1546] 6. Cost reduction proposals

[1547] The server generates specific cost-cutting proposals based on the identified waste, simulates the effectiveness of the proposals, and sets priorities.

[1548] Example: Propose specific measures to improve the energy efficiency of a server and list them in order of importance.

[1549] Additionally, the system of the present invention incorporating the emotion engine includes the following elements:

[1550] 1. Emotion recognition

[1551] The server uses an emotion engine to recognize the user's emotions and analyzes the user's writings and comments to identify the user's emotional state.

[1552] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[1553] 2. Emotion-based personalized recommendations

[1554] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. For example, if the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[1555] Example: When a user is feeling stressed, the server prioritizes suggestions for reducing stress that are relatively easy to implement.

[1556] 3. Emotional Feedback

[1557] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[1558] Example: The server displays a dashboard showing the user's current emotional state and provides suggestions for improvement based on that.

[1559] For example, if a company's human resources manager is feeling stressed, the system will recognize that emotion and first offer relatively simple and effective improvement suggestions, allowing the manager to start with suggestions that are easy to implement. The system also monitors the effectiveness of the suggestions after implementation and, if further suggestions are needed, suggests next steps based on the manager's emotional state.

[1560] In this way, the present invention aims to make corporate cost management more effective and user-friendly by combining AI technology and emotion recognition technology.

[1561] The processing flow will be explained below.

[1562] Step 1:

[1563] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data, which includes verifying credentials and specifying the time period to retrieve.

[1564] Example: A server retrieves last month's expense data from a company's ERP system via an API.

[1565] Step 2:

[1566] The server checks the integrity and completeness of the retrieved data, adapts to the data format and content, identifies missing or outlier values, and imputes data as needed.

[1567] Example: Standardizing the date format and number of digits in data collected by the server, and detecting and correcting inappropriate values.

[1568] Step 3:

[1569] The server performs basic statistical processing on the cleansed data to calculate the mean, median, and standard deviation of expenses.

[1570] Example: Calculating the average consumption and variance of energy consumption data collected by a server.

[1571] Step 4:

[1572] The server performs trend analysis based on past data and generates future cost forecasts, using AI algorithms to take into account seasonal fluctuations and the impact of economic conditions.

[1573] Example: A server predicts consumption trends for the next year based on energy data from the past 12 months.

[1574] Step 5:

[1575] The server automatically categorizes expense data by category, allows users to perform detailed analysis for each specific category, and allows users to manually adjust the categorization results as needed.

[1576] Example: A server categorizes expenses by department and project, clearly showing the breakdown of each.

[1577] Step 6:

[1578] The server uses AI technology to identify outliers, using predictive algorithms to spot spending items that fall outside of the normal range.

[1579] Example: A server detects extremely high communication costs and analyzes the background.

[1580] Step 7:

[1581] The server generates specific cost-saving proposals based on the identified outliers, simulates the effects of the proposals, and provides the proposals in an easy-to-implement format for the user.

[1582] Example: Propose specific measures to improve the energy efficiency of servers.

[1583] Step 8:

[1584] The server prioritizes the generated reduction proposals, allowing the user to implement the most effective proposals first.

[1585] Example: A server creates a prioritized list of energy usage optimization measures based on importance and urgency.

[1586] Step 9:

[1587] The server creates a detailed implementation plan based on the reduction proposals, including specific steps and schedules, which the user can review and adjust.

[1588] Example: A server creates an energy reduction plan, which the user reviews and adjusts to suit office hours.

[1589] Step 10:

[1590] The terminal executes cost-cutting activities based on the established plan, and the server monitors the progress in real time and provides feedback to the user.

[1591] Example: A user implements new energy usage guidelines, and a server monitors their progress and effectiveness.

[1592] Step 11:

[1593] The server then re-analyzes the post-execution data to confirm the effectiveness of the improvements and, if necessary, makes further optimization suggestions.

[1594] Example: A server checks the effectiveness of new energy usage methods and suggests further improvements.

[1595] (A system incorporating an emotion engine)

[1596] Step 12:

[1597] The server uses an emotion engine to recognize the user's emotions by analyzing the user's posts and comments to identify their emotional state.

[1598] Example: A server analyzes a user's chat messages and determines that the user is stressed.

[1599] Step 13:

[1600] The server customizes the content and priorities of cost-saving proposals based on the user's recognized emotions, giving priority to proposals that are less burdensome, taking into account the user's emotional state.

[1601] Example: When a server detects a user is feeling stressed, it prioritizes suggestions for reducing stress that are relatively easy to implement.

[1602] Step 14:

[1603] The server visually feeds back the user's emotional state, allowing the user to select suggestions that correspond to that state.

[1604] Example: The server displays a dashboard showing the user's current emotional state and provides appropriate improvement suggestions based on that.

[1605] In this way, a system that combines an emotion engine can make flexible cost optimization proposals that take into account the user's emotional state, aiming to make corporate cost management more efficient and user-friendly.

[1606] Example 2

[1607] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1608] Conventional cost management systems collect and analyze expense data, generate cost-reduction proposals, and formulate implementation plans, but they are unable to provide effective proposals that take into account the user's emotional state. This can lead to stress and difficulty in implementing the proposals. Therefore, there is a need for a system that recognizes the user's emotional state and provides customized proposals based on that state.

[1609] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1610] In this invention, the server includes means for automatically collecting expense data from data sources, means for checking the consistency and completeness of the collected expense data and supplementing it as necessary, means for analyzing the collected and supplemented data and identifying outliers, means for generating cost-cutting proposals based on the identified outliers, means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness, means for identifying the user's emotional state using an emotion engine that recognizes the user's emotions, means for customizing the content and priorities of the cost-cutting proposals based on the user's emotional state, and means for visually providing feedback on the user's emotional state, thereby enabling effective and less burdensome cost-cutting proposals to be made in accordance with the user's emotional state.

[1611] "Data sources" refers to the systems and applications that store a company's expense data, including ERP systems and accounting software.

[1612] "Expense data" is data that contains detailed information about the costs a company incurs on a day-to-day basis.

[1613] "Integrity" means that data is accurate, consistent, and does not contradict each other.

[1614] "Complete" means that the data contains all necessary information and is free of missing parts.

[1615] "Imputation" refers to filling in missing information in data using prediction or imputation algorithms.

[1616] An "outlier" is a value that is extremely outlier compared to other values ​​in a data set, and often includes errors or exceptions.

[1617] A "cost reduction proposal" refers to a proposal that outlines specific action items and strategies for reducing a company's expenses.

[1618] "Plan" refers to a document that contains specific steps and timelines for implementing the generated cost reduction proposals.

[1619] "Progress" refers to the status of the implementation of planned specific actions, and indicates how much progress has been made over time.

[1620] "Effects" refer to the results obtained as a result of implementing a plan, and specifically include the degree of cost reduction and improved efficiency.

[1621] "Emotion engine" refers to software equipped with artificial intelligence technology to recognize and analyze a user's emotional state.

[1622] "Emotional state" refers to the psychological state such as stress or satisfaction felt by the user.

[1623] "Feedback" refers to providing the analysis results and suggestions to the user visually or in other ways, allowing the user to check their own status and the suggestions.

[1624] This invention is a system that utilizes AI technology to improve the efficiency of corporate cost management, and is characterized by its incorporation of an emotion engine that recognizes user emotions. To implement this system, hardware and software including the following elements are used.

[1625] 1. Data Collection

[1626] The server accesses data sources such as ERP systems or accounting software and sends API requests to retrieve the company's expense data. Specifically, the server sends an HTTP request and receives data in JSON format from the ERP system or accounting software.

[1627] Example: A server sends a request to "https: / / api.erp.example.com / v1 / expenses?month=last" to retrieve last month's expense data from a company's ERP system.

[1628] 2. Data cleansing

[1629] The server checks the integrity and completeness of the data retrieved, imputes missing values ​​using predictive algorithms, and standardizes formats, including detecting duplicate entries in the database and using a reverse dictionary to infer missing dates.

[1630] Example: A server corrects inconsistencies in date formats between expense data entries and imputes missing date fields with a predictive algorithm.

[1631] 3. Cost Analysis

[1632] The server performs statistical processing to identify outliers in expenses and forecast future costs. Specifically, it uses the Pandas library to detect outliers and the Prophet model to forecast future cost trends.

[1633] Example: A server analyzes energy consumption data and predicts energy consumption trends for the next year.

[1634] 4. Categorization

[1635] The server automatically categorizes expense data into categories and allows users to review and adjust the details, including using machine learning models to categorize data into specific categories.

[1636] Example: A server organizes expense data into categories such as communication expenses, travel expenses, and advertising expenses.

[1637] 5. Identify waste

[1638] The server uses anomaly detection algorithms to identify wasteful costs and evaluate potential areas for improvement, using techniques such as Isolation Forest and One-Class SVM.

[1639] Example: The server identifies communication expenses that are significantly higher than average as an anomaly and analyzes the details.

[1640] 6. Cost reduction proposals

[1641] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[1642] Example: A server generates a list of proposals for replacing lighting with LEDs to improve energy efficiency, based on estimated results.

[1643] 7. Emotion recognition

[1644] The server uses an emotion engine to recognize user emotions and identify the emotional state from user posts and statements, using a natural language processing model (e.g., BERT).

[1645] Example: A server analyzes a user's message "Work has been very stressful lately" and determines that the user is feeling stressed.

[1646] 8. Emotion-based personalized recommendations

[1647] The server customizes the content and priority of cost-cutting proposals according to the user's emotions. If the user is feeling stressed, it will prioritize proposals that put less strain on the user.

[1648] Example: When a user feels stressed, the server displays a low-impact reduction suggestion such as "reduce the frequency of meetings."

[1649] 9. Emotional Feedback

[1650] The server provides visual feedback of the user's emotional state, allowing the user to check the state and select suggestions using a visualization tool.

[1651] Example: The server displays the user's emotional state on a dashboard as "Stress level: High" and provides suggestions for improvement based on that.

[1652] This system allows companies to achieve efficient and user-friendly cost management, allowing users to receive cost-saving suggestions that take their emotional state into account and are easy to implement.

[1653] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1654] Program processing flow

[1655] Step 1: Data collection

[1656] The server contacts the data source and sends an API request to retrieve expense data.

[1657] Input: API endpoint URL and authentication information configured on the server.

[1658] Specific behavior: The server sends an HTTP request to an endpoint such as "https: / / api.erp.example.com / v1 / expenses?month=last".

[1659] Output: Expense data in JSON format returned from your ERP system or accounting software.

[1660] Step 2: Data cleansing

[1661] The server checks the consistency and completeness of the data it retrieves, fills in missing values, and standardizes the format.

[1662] Input: Captured expense data in JSON format.

[1663] What it does: The server uses a database library to read data, detects duplicate entries, fills in missing data with a predictive algorithm, and standardizes data formats (for example, standardizing date formats to "YYYY-MM-DD").

[1664] Output: A clean dataset that has been checked for consistency and completeness, imputed for missing values, and has a uniform format.

[1665] Step 3: Cost analysis

[1666] The server analyzes the cleansed data to identify outliers and predict future costs.

[1667] Input: The cleansed dataset.

[1668] What it does: The server uses a statistical library to detect outliers in the data and a time series forecasting model (e.g., Prophet) to predict future cost trends.

[1669] Output: Identified outliers and future cost forecast data.

[1670] Step 4: Categorization

[1671] The server automatically categorizes expense data and allows users to review and adjust the details.

[1672] Input: The cleansed dataset.

[1673] What it does: The server uses machine learning models to categorize expense data by category (e.g., communication, travel, advertising).

[1674] Output: Expense data broken down by category.

[1675] Step 5: Identify waste

[1676] The server uses an anomaly detection algorithm to identify wasteful costs and evaluate potential for improvement.

[1677] Input: Expense data broken down by category.

[1678] How it works: The server uses outlier detection algorithms such as Isolation Forest and One-Class SVM to identify unreasonably high expenditures.

[1679] Output: Identified waste costs and a detailed report.

[1680] Step 6: Cost reduction proposal

[1681] The server generates specific cost-saving proposals based on the identified waste, and evaluates and prioritizes the proposals using heuristic rules and simulation models.

[1682] Input: Detailed data on waste costs.

[1683] Specific operation: The server generates improvement measures using heuristic rules and verifies their effectiveness using a simulation model.

[1684] Output: Prioritized cost-saving proposals and their predicted effectiveness.

[1685] Step 7: Emotion Recognition

[1686] The server uses an emotion engine to recognize the user's emotions and analyzes the user's emotional state from their statements and writings.

[1687] Input: User message and feedback data.

[1688] What happens: The server analyzes the text data using a natural language processing model (e.g., BERT) to identify the user's emotional state.

[1689] Output: User's emotional state report.

[1690] Step 8: Customize your suggestions based on emotions

[1691] The server customizes the suggestions and priorities based on the user's emotions recognized.

[1692] Input: User emotional state report, cost reduction suggestions.

[1693] Specific operation: The server selects and customizes suggestions that are easy to implement and have high priority based on the user's emotional state.

[1694] Output: Customized cost-saving proposals.

[1695] Step 9: Emotional Feedback

[1696] The server visually feeds back the user's emotional state, allowing the user to confirm the state and select suggestions.

[1697] Input: customized cost-saving suggestions, user emotional state report.

[1698] Specific operation: The server uses visualization tools such as a dashboard to display the user's emotional state and suggestions.

[1699] Output: Emotional feedback and improvement suggestions on a visualized dashboard.

[1700] (Application example 2)

[1701] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1702] Conventional cost management systems collect and analyze expense data, identify outliers, and generate cost-reduction proposals. However, these processes do not take into account the user's emotional state, often resulting in stress. Furthermore, in logistics centers, worker comfort and efficiency are important, and workload adjustments based on individual emotional states are required. However, current systems lack an effective means of doing this, resulting in insufficient improvements in work efficiency and reductions in worker stress.

[1703] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for automatically collecting expense data from data sources; means for verifying the consistency and completeness of the collected expense data and supplementing it as necessary; means for analyzing the collected and supplemented data and identifying outliers; means for generating cost reduction proposals based on the identified outliers; means for formulating a plan for implementing the generated proposals and monitoring their progress and effectiveness; means for determining the user's emotional state using emotion recognition technology and customizing the content and priorities of the proposals based on the determined emotional state; and means having a human interface for presenting the customized proposals to the user. This enables efficient cost management that takes the user's emotional state into consideration, reducing worker stress, and improving work efficiency at logistics centers.

[1704] "Data sources" refers to information sources such as ERP systems or accounting software that store a company's expense data.

[1705] "Expense data" is detailed information about the costs a company incurs on a daily basis, including travel, communication, and energy costs.

[1706] "Integrity" refers to data that is consistently accurate and consistent across different data sets.

[1707] "Complete" refers to the state in which all data is available and there are no missing values ​​or incomplete information.

[1708] "Imputation" refers to the process of estimating and adding missing data to complete a dataset.

[1709] An "outlier" is a data value that deviates from normal patterns or acceptable ranges.

[1710] "Cost reduction proposals" refer to specific action plans and measures to reduce a company's expenses.

[1711] A "plan" is a schedule that defines the steps and resources needed to achieve a specific goal.

[1712] "Progress" refers to the status of execution and degree of success of a plan or project.

[1713] "Effect" refers to the results or outcomes that result from a certain action or measure.

[1714] "Emotion recognition technology" refers to technology that identifies a user's emotional state (e.g., stress, joy, anger, etc.) from facial expressions and vocal tone.

[1715] "User" refers to the person who operates the system and manages and analyzes expense data.

[1716] "Emotional state" refers to a user's current psychological state as determined by emotion recognition technology.

[1717] "Human interface" refers to the interface through which a user interacts with a system and exchanges information.

[1718] This invention provides a system aimed at efficient cost management and worker stress reduction in logistics centers. This system utilizes AI technology and emotion recognition technology to make suggestions based on the user's emotional state. Specific embodiments are described below.

[1719] The system consists of a server and smart glasses as its main components. The server automatically collects expense data from data sources such as a company's ERP system or accounting software, including the means to retrieve the required data using API requests.

[1720] The server checks the consistency and completeness of collected expense data, fills in missing values ​​with a predictive algorithm, and standardizes the format, thereby ensuring data quality.

[1721] The server then performs statistical processing and analyses on the collected and stored data, identifying outliers and conducting trend analysis to predict future costs, which in turn generates cost-saving recommendations.

[1722] The server uses emotion recognition technology to collect data through the smart glasses to determine the worker's emotional state. Specifically, it uses facial recognition and voice analysis to determine the worker's emotional state from their facial expressions and tone. Based on the determined emotional state, the server customizes the content and priorities of suggestions.

[1723] The customized suggestions are sent from the server to the smart glasses and presented to the worker through a human interface, so that suggestions for a lighter workload are displayed preferentially to workers who are feeling stressed.

[1724] The server uses TensorFlow and OpenCV to build emotion recognition models, Flask framework to manage data on the server side, and SQLite to operate a local database. Communication with the smart glasses is via Bluetooth Low Energy (BLE).

[1725] For example, if a worker at a logistics center is feeling stressed, the system will recognize that emotion and suggest lighter tasks first, such as avoiding carrying heavy items and prioritizing stocking lighter items, thereby reducing the worker's stress and allowing them to work efficiently.

[1726] Below is an example of a prompt sentence that uses a generative AI model to design suggestions based on the worker's emotions.

[1727] Example prompt sentence:

[1728] Consider a graphical interface design that uses an emotion engine to identify the emotions of workers and suggest lighter tasks to stressed workers. In particular, be sure to consider the use of colors and layout to reduce stress.

[1729] In this way, this system simultaneously improves the cost performance and work efficiency of logistics centers through efficient management of expense data and suggestions based on the emotional state of workers.

[1730] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1731] Step 1:

[1732] The server collects expense data from data sources, specifically, from ERP systems and accounting software using API requests. The input is the access information of the data source based on the API request, and the output is the collected raw expense data.

[1733] Step 2:

[1734] The server checks the collected expense data for consistency and completeness and imputes it if necessary, correcting inconsistent date formats and imputing missing values ​​using predictive algorithms. The input is the raw expense data collected, and the output is cleaned data that has been checked for consistency and completeness.

[1735] Step 3:

[1736] The server analyzes the collected and imputed data to identify outliers. It performs statistical processing to detect outliers that deviate from the normal range. The input is the cleaned expense data, and the output is a list of outliers.

[1737] Step 4:

[1738] The server generates cost reduction proposals based on the identified outliers. It analyzes the causes of the outliers and proposes specific cost reduction measures accordingly. The input is a list of outliers, and the output is the generated cost reduction proposals.

[1739] Step 5:

[1740] The server uses emotion recognition technology to determine the worker's emotional state. It uses the camera and microphone in the smart glasses to perform facial recognition and voice analysis to identify the worker's emotional state. The input is real-time video and audio data, and the output is the determined emotional state.

[1741] Step 6:

[1742] The server customizes the content and priority of the proposals based on the determined emotional state. If the user is feeling stressed, it prioritizes proposals that are less stressful. The input is the determined emotional state and the generated cost-saving proposals, and the output is the customized proposals.

[1743] Step 7:

[1744] The server sends the customized suggestions to the smart glasses for presentation to the worker. The suggestions are displayed to the worker via a human interface using Bluetooth Low Energy (BLE). The input is the customized suggestions, and the output is the suggestions displayed on the smart glasses display.

[1745] Step 8:

[1746] The server develops an implementation plan for the proposals and monitors their progress and effectiveness. It evaluates the results of implementing the proposed cost-saving measures and obtains feedback. The inputs are the implemented proposals and their results data, and the outputs are progress reports and effectiveness evaluation reports.

[1747] By carrying out the above processing steps, efficient cost management and stress reduction for workers at the logistics center can be achieved.

[1748] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1749] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1750] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1751] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1752] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1753] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1754] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1755] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1756] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1757] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1758] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1759] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1760] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1761] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1762] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1763] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1764] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1765] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1766] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1767] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1768] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1769] The following is further disclosed regarding the above embodiment.

[1770] (Claim 1)

[1771] A means of automatically collecting expense data from data sources; and

[1772] A means to verify the consistency and completeness of collected expense data and supplement it as necessary;

[1773] A means of analyzing the collected and imputed data and identifying outliers;

[1774] means for generating cost reduction recommendations based on the identified outliers;

[1775] Developing plans for implementing the generated recommendations and means of monitoring their progress and effectiveness;

[1776] A system including:

[1777] (Claim 2)

[1778] 10. The system of claim 1, further comprising means for categorizing the expense data and performing detailed analysis based on specific categories.

[1779] (Claim 3)

[1780] 10. The system of claim 1, further comprising means for prioritizing the reduction proposals and developing an implementation plan based on the priorities.

[1781] "Example 1"

[1782] (Claim 1)

[1783] A means of automatically collecting expense data from data sources; and

[1784] A means to check the consistency and completeness of collected expense data and impute missing values ​​with predictive models;

[1785] A means for performing statistical processing on the collected and supplemented data to identify outliers;

[1786] a means for generating cost reduction recommendations using an AI algorithm based on the identified anomalies;

[1787] A means for simulating the effects of generated proposals in advance and setting priorities;

[1788] A means to develop action plans based on priorities and monitor their progress and effectiveness in real time;

[1789] A system including:

[1790] (Claim 2)

[1791] 10. The system of claim 1, further comprising means for categorizing the expense data by department or project and performing detailed analysis based on the particular category.

[1792] (Claim 3)

[1793] 2. The system according to claim 1, further comprising means for simulating the effects of reduction proposals in advance and prioritizing them based on the effects.

[1794] "Application Example 1"

[1795] (Claim 1)

[1796] A means of automatically collecting expense data from data sources; and

[1797] A means to verify the consistency and completeness of collected expense data and supplement it as necessary;

[1798] A means of analyzing the collected and imputed data and identifying outliers;

[1799] means for generating cost reduction recommendations based on the identified outliers;

[1800] Developing plans for implementing the generated recommendations and means of monitoring their progress and effectiveness;

[1801] Through a smartphone application, you can grasp the resource usage status of the logistics center in real time, and quickly identify and reduce unnecessary costs.

[1802] A system including:

[1803] (Claim 2)

[1804] 10. The system of claim 1, further comprising means for categorizing the expense data and performing detailed analysis based on specific categories.

[1805] (Claim 3)

[1806] 10. The system of claim 1, further comprising means for prioritizing the reduction proposals and developing an implementation plan based on the priorities.

[1807] "Example 2: Combining Emotion Engines"

[1808] (Claim 1)

[1809] A means of automatically collecting expense data from data sources; and

[1810] A means to verify the consistency and completeness of collected expense data and supplement it as necessary;

[1811] A means of analyzing the collected and imputed data and identifying outliers;

[1812] means for generating cost reduction recommendations based on the identified outliers;

[1813] Developing plans for implementing the generated recommendations and means of monitoring their progress and effectiveness;

[1814] means for identifying an emotional state using an emotion engine that recognizes the user's emotions;

[1815] a means for customizing the content and prioritization of cost-saving suggestions based on the user's emotional state;

[1816] A system including a means for providing visual feedback of a user's emotional state.

[1817] (Claim 2)

[1818] 10. The system of claim 1, further comprising means for categorizing the expense data and performing detailed analysis based on specific categories.

[1819] (Claim 3)

[1820] 10. The system of claim 1, further comprising means for prioritizing the reduction proposals and developing an implementation plan based on the priorities.

[1821] "Application example 2 when combining emotion engines"

[1822] (Claim 1)

[1823] A means of automatically collecting expense data from data sources; and

[1824] A means to verify the consistency and completeness of collected expense data and supplement it as necessary;

[1825] A means of analyzing the collected and imputed data and identifying outliers;

[1826] means for generating cost reduction recommendations based on the identified outliers;

[1827] Developing plans for implementing the generated recommendations and means of monitoring their progress and effectiveness;

[1828] Determine the user's emotional state using emotion recognition technology;

[1829] a means for customizing the content and prioritization of suggestions based on the determined emotional state; and

[1830] means with a human interface for presenting customized suggestions to a user;

[1831] A system including:

[1832] (Claim 2)

[1833] 10. The system of claim 1, further comprising means for categorizing the expense data and performing detailed analysis based on specific categories.

[1834] (Claim 3)

[1835] 10. The system of claim 1, further comprising means for prioritizing the reduction proposals and developing an implementation plan based on the priorities. [Explanation of symbols]

[1836] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of automatically collecting expense data from data sources; and A means to verify the consistency and completeness of collected expense data and supplement it as necessary; A means of analyzing the collected and imputed data and identifying outliers; means for generating cost reduction recommendations based on the identified outliers; Developing plans for implementing the generated recommendations and means of monitoring their progress and effectiveness; A system including:

2. 10. The system of claim 1, further comprising means for categorizing expense data and performing detailed analysis based on specific categories.

3. 10. The system of claim 1, further comprising means for prioritizing the reduction proposals and developing an implementation plan based on the priorities.

Images