system

Abstract

The system according to this embodiment aims to reduce errors caused by the processing of non-standard invoices and manual work, thereby achieving efficient invoice processing. [Solution] The system according to the embodiment comprises a data extraction unit, a contract information matching unit, an anomaly detection unit, a payment schedule optimization unit, and a claim handling unit. The data extraction unit extracts data from invoices. The contract information matching unit matches the data extracted by the data extraction unit with contract information and transaction history. The anomaly detection unit detects anomalies based on the results of the matching by the contract information matching unit. The payment schedule optimization unit optimizes the payment schedule based on the anomalies detected by the anomaly detection unit. The claim handling unit handles claims based on the schedule optimized by the payment schedule optimization unit.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, there is a problem that it is easy to generate non-standard claim processing and manual errors, and it is difficult to perform efficient claim processing.

[0005] The system according to the embodiment aims to reduce non-standard claim processing and manual errors and realize efficient claim processing.

Means for Solving the Problems

[0006] The system according to the embodiment comprises a data extraction unit, a contract information matching unit, an anomaly detection unit, a payment schedule optimization unit, and a claims handling unit. The data extraction unit extracts data from invoices. The contract information matching unit matches the data extracted by the data extraction unit with contract information and transaction history. The anomaly detection unit detects anomalies based on the results of the matching by the contract information matching unit. The payment schedule optimization unit optimizes the payment schedule based on the anomalies detected by the anomaly detection unit. The claims handling unit handles claims based on the schedule optimized by the payment schedule optimization unit. [Effects of the Invention]

[0007] The system according to this embodiment can reduce errors caused by manual processing and the handling of non-standard invoices, thereby achieving efficient invoice processing. [Brief explanation of the drawing]

[0008] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0009] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0010] First, let's explain the terminology used in the following explanation.

[0011] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), or TPU (Tensor Processing Unit).

[0012] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0013] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0014] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F manages communication between multiple computers. Examples of communication standards applicable to the communication I / F include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

[0016] [First Embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] As shown in FIG. 1, the 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.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. 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. Also, the database 24 and the communication I / F 26 are 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).

[0019] The smart device 14 comprises a computer 36, a receiving device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The receiving device 38, output device 40, and camera 42 are also connected to the bus 52.

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0023] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0025] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0026] In the smart device 14, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0027] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example of form 1) The self-learning generation AI agent system, specialized for invoice processing and payment operations according to an embodiment of the present invention, is a system that extracts invoice data and enables real-time matching with contract information and transaction history. This system utilizes AIOCR to extract invoice data and enables real-time matching with contract information and transaction history. This provides a comprehensive solution for detecting inconsistencies, notifying anomalies, automatically creating optimal payment schedules, and handling claims quickly. Multiple AI agents cooperate to carry out tasks, immediately proposing solutions when problems occur, thereby streamlining the entire operation and improving reliability. For example, it starts with receiving invoices. For example, it collects invoices in paper or electronic format and handles various formats by utilizing the data classification capabilities of the generation AI agent. Next, it extracts invoice data (invoice amount, date, billing source, etc.) using high-precision AIOCR, and reduces misreadings and omissions by utilizing the learning capabilities of the generation AI agent. The extracted data is matched with contract information and transaction history in real time, and inconsistencies are detected. Rapid data integration is achieved by utilizing the knowledge graph of the generation AI agent. Furthermore, anomalies and fraudulent billing are detected by comparing with past data. Immediate identification of fraudulent billing is achieved by utilizing the anomaly detection algorithm of the generation AI agent. The system creates payment schedules considering cash flow and supplier priorities, and leverages the decision-making support capabilities of the generating AI agent to achieve optimal payment plans. If problems arise, it notifies relevant departments and suppliers and proposes solutions. Processing results are recorded to aid in future business improvements. The system utilizes the language generation capabilities of the generating AI agent to enable rapid and accurate complaint handling, and leverages its analytical capabilities to achieve continuous business improvement. In this way, it provides a comprehensive solution to streamline invoice processing and payment operations and improve reliability. This self-learning generating AI agent system, specialized for invoice processing and payment operations, can streamline and improve the reliability of these processes.

[0029] The self-learning generation AI agent system, specialized for invoice processing and payment operations according to the embodiment, comprises a data extraction unit, a contract information matching unit, an anomaly detection unit, a payment schedule optimization unit, and a claims handling unit. The data extraction unit extracts data from invoices. The data extraction unit extracts invoice data using, for example, high-precision AIOCR. The data extraction unit can improve accuracy by applying different extraction algorithms depending on the invoice format. The data extraction unit can determine importance based on the content of the invoice and select data to be extracted preferentially. The contract information matching unit matches the data extracted by the data extraction unit with contract information and transaction history. The contract information matching unit matches invoice data with contract information and transaction history in real time. The contract information matching unit can adjust the timing of matching based on the update frequency of contract information and transaction history. The contract information matching unit can apply different matching algorithms depending on the importance of the contract information. The anomaly detection unit detects anomalies based on the results matched by the contract information matching unit. The anomaly detection unit detects abnormal values ​​and fraudulent invoices by comparing them with past data. The anomaly detection unit can, for example, adjust the accuracy of its detection by considering the reliability of the trading partner when an anomaly is detected. The anomaly detection unit can, for example, adjust the detection criteria by referring to relevant laws and regulations when an anomaly is detected. The payment schedule optimization unit optimizes the payment schedule based on the anomalies detected by the anomaly detection unit. The payment schedule optimization unit creates a payment schedule based on, for example, cash flow and trading partner priority. The payment schedule optimization unit can dynamically adjust the payment schedule based on, for example, fluctuations in cash flow. The payment schedule optimization unit can optimize the payment schedule based on, for example, the credit information of the trading partner. The claims handling unit handles claims based on the schedule optimized by the payment schedule optimization unit. The claims handling unit notifies relevant departments and trading partners when a problem occurs and proposes a solution. The claims handling unit can, for example, select the optimal handling method by referring to past claims data when handling a claim.The claims handling unit can, for example, apply different response algorithms depending on the type of claim when handling a claim. This allows the self-learning, generation-type AI agent system, specialized for invoice processing and payment operations according to the embodiment, to streamline and improve the reliability of invoice processing and payment operations.

[0030] The data extraction unit extracts data from invoices. For example, the data extraction unit uses high-precision AIOCR to extract invoice data. Specifically, AIOCR is a technology that analyzes image data of invoices and converts it into text data. This allows for the accurate extraction of handwritten or printed invoice contents as digital data. The data extraction unit can improve accuracy by applying different extraction algorithms depending on the invoice format. Since invoice formats vary widely and differ from company to company, the data extraction unit selects the optimal algorithm based on pre-learned format information. For example, applying an algorithm specialized for a particular company's invoice format can improve extraction accuracy. The data extraction unit can also determine importance based on the invoice content and select data to be extracted preferentially. For example, it can prioritize the extraction of important data such as the invoice amount, payment deadline, and customer information, and manage it separately from other data. This allows the data extraction unit to quickly and accurately extract important information from invoices and utilize it for subsequent processing. Furthermore, the data extraction unit centrally manages the extracted data and can collaborate with other systems and departments as needed. For example, extracted data is stored on a cloud server, making it accessible to the contract information matching unit and the anomaly detection unit. Furthermore, by adjusting the frequency and accuracy of data extraction, flexible responses to specific situations and conditions become possible. This allows the data extraction unit to extract data efficiently and effectively, improving the overall system performance.

[0031] The contract information matching unit matches the data extracted by the data extraction unit with contract information and transaction history. For example, the contract information matching unit matches invoice data with contract information and transaction history in real time. Specifically, the contract information matching unit verifies whether the amount and transaction details stated on the invoice match the pre-registered contract information and past transaction history. This helps prevent fraudulent and erroneous billing. The contract information matching unit can adjust the timing of matching based on the update frequency of contract information and transaction history. For example, if contract information is frequently updated, the matching timing can be shortened to perform matching based on the latest information. Also, if there is a large transaction history, past data can be efficiently searched and matching can be performed quickly. The contract information matching unit can apply different matching algorithms depending on the importance of the contract information. For example, a more rigorous matching algorithm can be applied to high-value transactions and important contracts to improve accuracy. On the other hand, a simpler algorithm can be applied to relatively small transactions and general contracts to perform quick matching. In this way, the contract information matching unit can match data efficiently and accurately, improving the reliability of the entire system. Furthermore, the contract information verification unit can notify other departments and systems of the verification results in real time, supporting a rapid response. For example, if an anomaly is detected, it immediately notifies the anomaly detection unit, allowing appropriate countermeasures to be taken. In addition, based on the verification results, the payment schedule optimization unit provides information to create an optimal payment schedule. In this way, the contract information verification unit can improve the overall efficiency and reliability of the system.

[0032] The anomaly detection unit detects anomalies based on the results verified by the contract information verification unit. For example, the anomaly detection unit detects abnormal values ​​and fraudulent invoices by comparing them with past data. Specifically, the anomaly detection unit learns normal transaction patterns based on past transaction data and invoice data, and detects abnormal transactions based on that. For example, if there is an invoice that deviates significantly from the normal transaction amount or frequency, it is detected as an anomaly. For example, when detecting an anomaly, the anomaly detection unit can adjust the accuracy of detection by considering the reliability of the trading partner. For example, it can loosen the anomaly detection threshold for invoices from reliable trading partners and set a stricter threshold for invoices from unreliable trading partners. This improves the accuracy of anomaly detection. For example, when detecting an anomaly, the anomaly detection unit can adjust the detection criteria by referring to relevant laws and regulations. For example, it can set anomaly detection criteria based on laws and regulations in a specific industry or region and detect transactions that may violate those laws and regulations. This allows the anomaly detection unit to perform anomaly detection in compliance with laws and regulations. Furthermore, the anomaly detection unit provides information to respond quickly to detected anomalies. For example, if an anomaly is detected, the anomaly detection unit notifies the claims handling department of its detailed information to support appropriate action. Furthermore, the anomaly detection unit manages the history of detected anomalies, which can be used to improve the accuracy of future anomaly detection. This allows the anomaly detection unit to improve the overall reliability and safety of the system.

[0033] The payment schedule optimization unit optimizes the payment schedule based on anomalies detected by the anomaly detection unit. For example, the payment schedule optimization unit creates a payment schedule based on cash flow and supplier priorities. Specifically, the payment schedule optimization unit monitors the company's cash flow situation in real time and calculates the optimal payment timing. This optimizes the company's cash flow and reduces unnecessary spending. The payment schedule optimization unit can dynamically adjust the payment schedule based on fluctuations in cash flow. For example, it can adjust the payment schedule to coincide with periods when cash flow is prone to fluctuations, such as the end of the month or quarter, to stabilize cash flow. It can also optimize the payment schedule based on the credit information of suppliers. For example, it can make early payments to suppliers with high credit scores and delay payments to suppliers with low credit scores to mitigate risk. This allows the payment schedule optimization unit to optimize the company's cash flow and maintain good relationships with suppliers. Furthermore, the payment schedule optimization unit also responds appropriately to anomalies detected by the anomaly detection unit. For example, if fraudulent billing is detected, it will withhold payment and conduct a detailed investigation. Furthermore, once the anomaly is resolved, the payment schedule will be promptly readjusted and appropriate measures will be taken with the business partner. This allows the payment schedule optimization unit to respond flexibly and quickly to anomalies, thereby improving the company's reliability.

[0034] The Claims Handling Department handles claims based on a schedule optimized by the Payment Schedule Optimization Department. For example, when a problem occurs, the Claims Handling Department notifies relevant departments and business partners and proposes solutions. Specifically, when a problem occurs, the Claims Handling Department promptly notifies relevant departments and business partners and provides detailed information about the problem. This allows stakeholders to quickly begin taking action and resolve the problem early. For example, when handling a claim, the Claims Handling Department can select the optimal response method by referring to past claim data. For example, based on the response methods and results from similar problems in the past, it can select the optimal solution and resolve the problem quickly and effectively. Furthermore, the Claims Handling Department can apply different response algorithms depending on the type of claim. For example, for claims regarding payment delays, it readjusts the payment schedule, and for quality-related claims, it collaborates with the quality control department to implement countermeasures. This allows the Claims Handling Department to handle various types of claims appropriately and maintain trust with business partners. In addition, the Claims Handling Department can manage the history of claim handling and use it to improve the accuracy of future claim handling. For example, by analyzing past complaint handling history and identifying common problems and areas for improvement, the quality of future complaint handling can be enhanced. Furthermore, the complaint handling department can provide feedback on the results of complaint handling to relevant departments, which can be used to improve business processes. This allows the complaint handling department to improve the overall reliability and efficiency of the system.

[0035] The data extraction unit can extract invoice data using high-precision optical character recognition (OCR) technology. The data extraction unit can, for example, extract invoice data using high-precision optical character recognition technology. The data extraction unit can, for example, extract invoice data using an OCR engine. The data extraction unit can, for example, extract invoice data using techniques for improving accuracy. As a result, the accuracy of invoice data extraction is improved by using high-precision optical character recognition technology.

[0036] The contract information matching unit can match invoice data with contract information and transaction history in real time. For example, the contract information matching unit matches invoice data with contract information and transaction history in real time. The contract information matching unit can achieve real-time matching by, for example, increasing the matching frequency. The contract information matching unit can also achieve real-time matching by, for example, reducing the delay time. This enables rapid detection of discrepancies through real-time matching.

[0037] The anomaly detection unit can detect abnormal values ​​and fraudulent claims by comparing them with past data. For example, the anomaly detection unit can detect abnormal values ​​by comparing them with past data. For example, the anomaly detection unit can detect fraudulent claims by comparing them with past data. For example, the anomaly detection unit can detect abnormal values ​​by setting an abnormal value threshold. For example, the anomaly detection unit can detect fraudulent claims by analyzing patterns of fraudulent claims. This improves the accuracy of detecting abnormal values ​​and fraudulent claims by comparing them with past data.

[0038] The payment schedule optimization unit can create payment schedules based on cash flow and customer priority. For example, the payment schedule optimization unit can create payment schedules based on cash flow. For example, the payment schedule optimization unit can create payment schedules based on customer priority. For example, the payment schedule optimization unit can create payment schedules using optimization algorithms. For example, the payment schedule optimization unit can create payment schedules by setting the elements to be considered. This achieves an optimal payment schedule by considering cash flow and customer priority.

[0039] The complaints handling department can notify relevant departments and business partners when a problem occurs and propose solutions. For example, the complaints handling department can notify relevant departments when a problem occurs. For example, the complaints handling department can notify business partners when a problem occurs. For example, the complaints handling department can resolve problems by proposing solutions. For example, the complaints handling department can set up notification methods to notify relevant departments and business partners. This improves the reliability of operations by enabling a swift response to problems and the proposal of solutions.

[0040] The data extraction unit can improve accuracy by applying different extraction algorithms based on the invoice format. For example, if the invoice is handwritten, the data extraction unit can apply a handwriting recognition algorithm to extract the data. For example, if the invoice is in PDF format, the data extraction unit can apply a PDF analysis algorithm to extract the data. For example, if the invoice is in image format, the data extraction unit can apply an image recognition algorithm to extract the data. In this way, the accuracy of data extraction is improved by applying an extraction algorithm according to the invoice format.

[0041] The data extraction unit can determine the importance of data based on the content of the invoices and select data to extract preferentially. For example, the data extraction unit can prioritize extracting invoices with high invoice amounts. For example, the data extraction unit can prioritize extracting invoices with approaching payment deadlines. For example, the data extraction unit can prioritize extracting invoices from highly reliable trading partners. In this way, by determining the importance of data based on the content of the invoices, important data can be extracted preferentially.

[0042] The data extraction unit can apply different extraction methods based on the language and region of the invoice. For example, it can apply an English recognition algorithm to an English invoice, or a Japanese recognition algorithm to a Japanese invoice. For example, it can apply a region-specific extraction method to an invoice with a region-specific format. This improves the accuracy of data extraction by applying extraction methods tailored to the language and region of the invoice.

[0043] The data extraction unit can adjust the accuracy of data extraction based on the reliability of the invoice issuer. For example, the data extraction unit applies a high-precision extraction algorithm to invoices from reliable issuers. For example, the data extraction unit can carefully extract and verify data from invoices from unreliable issuers. For example, the data extraction unit can adjust the verification process of extracted data according to the reliability of the issuer. This improves the accuracy of data extraction by considering the reliability of the invoice issuer.

[0044] The contract information matching unit can adjust the matching timing based on the update frequency of contract information and transaction history. For example, if contract information is frequently updated, the contract information matching unit can shorten the matching timing. For example, if transaction history is frequently updated, the contract information matching unit can shorten the matching timing. For example, the contract information matching unit can dynamically adjust the matching timing according to the update frequency of contract information and transaction history. This improves the accuracy of matching by adjusting the matching timing based on the update frequency of contract information and transaction history.

[0045] The contract information matching unit can apply different matching algorithms based on the importance of the contract information during matching. For example, the contract information matching unit can apply a high-precision matching algorithm to contract information of high importance. For example, the contract information matching unit can apply a rapid matching algorithm to contract information of low importance. For example, the contract information matching unit can dynamically select a matching algorithm according to the importance of the contract information. This improves the accuracy of matching by applying a matching algorithm appropriate to the importance of the contract information.

[0046] The contract information matching unit can apply different matching methods based on the region and language of the contract information and transaction history. For example, the contract information matching unit can apply an English recognition algorithm to English contract information. For example, the contract information matching unit can apply a Japanese recognition algorithm to Japanese contract information. For example, the contract information matching unit can apply a region-specific matching method to contract information with a region-specific format. This improves the accuracy of matching by applying matching methods according to the region and language of the contract information and transaction history.

[0047] The contract information matching unit can adjust the accuracy of the matching based on the reliability of the contract information. For example, the contract information matching unit applies a high-precision matching algorithm to highly reliable contract information. For example, the contract information matching unit can perform careful matching and verification on less reliable contract information. For example, the contract information matching unit can dynamically adjust the matching accuracy according to the reliability of the contract information. This improves the accuracy of the matching by considering the reliability of the contract information.

[0048] The anomaly detection unit can adjust the detection accuracy based on the reliability of the trading partner when an anomaly is detected. For example, the anomaly detection unit applies a high-precision detection algorithm to data from highly reliable trading partners. For example, the anomaly detection unit can perform careful detection and verification on data from less reliable trading partners. For example, the anomaly detection unit can dynamically adjust the detection accuracy according to the reliability of the trading partner. This improves the accuracy of anomaly detection by considering the reliability of the trading partner.

[0049] The anomaly detection unit can adjust its detection criteria by referring to relevant laws and regulations when an anomaly is detected. For example, the anomaly detection unit sets the anomaly detection criteria based on laws and regulations. For example, the anomaly detection unit can dynamically adjust the anomaly detection criteria in response to changes in laws and regulations. For example, the anomaly detection unit can improve the accuracy of detection by referring to relevant laws and regulations when an anomaly is detected. This allows for appropriate adjustment of the anomaly detection criteria by referring to relevant laws and regulations.

[0050] The payment schedule optimization unit can dynamically adjust the payment schedule based on fluctuations in cash flow. For example, if cash flow deteriorates, the payment schedule optimization unit will review the payment schedule and postpone payments. For example, if cash flow improves, the payment schedule optimization unit can bring forward the payment schedule and make payments earlier. For example, the payment schedule optimization unit can dynamically adjust the payment schedule in response to fluctuations in cash flow to achieve an optimal payment plan. In this way, by adjusting the payment schedule in response to fluctuations in cash flow, an optimal payment plan is achieved.

[0051] The payment schedule optimization unit can adjust payment schedules based on the credit information of trading partners. For example, the payment schedule optimization unit can make payments early to trading partners with high credit scores. For example, the payment schedule optimization unit can delay payments to trading partners with low credit scores. For example, the payment schedule optimization unit can dynamically adjust payment schedules based on the credit information of trading partners to achieve an optimal payment plan. In this way, an optimal payment plan is achieved by optimizing payment schedules based on the credit information of trading partners.

[0052] The payment schedule optimization unit can adjust the payment schedule based on the geographical distribution of trading partners when creating the payment schedule. For example, if trading partners are concentrated in a nearby area, the payment schedule optimization unit can make payments in a batch. For example, if trading partners are scattered far apart, the payment schedule optimization unit can make payments individually. For example, the payment schedule optimization unit can dynamically adjust the payment schedule based on the geographical distribution of trading partners to achieve an optimal payment plan. In this way, an optimal payment plan is achieved by taking into account the geographical distribution of trading partners.

[0053] The payment schedule optimization unit can optimize payment schedules by referring to relevant market data when creating them. For example, the payment schedule optimization unit adjusts payment schedules based on market trends. For example, the payment schedule optimization unit can adjust payment schedules based on market interest rate trends. For example, the payment schedule optimization unit can dynamically adjust payment schedules based on market supply and demand balance to achieve an optimal payment plan. This allows for the realization of an optimal payment plan by referring to relevant market data.

[0054] The claims handling department can select the optimal response method when handling a claim by referring to past claims data. For example, the claims handling department can refer to the response methods used in similar past claims. For example, the claims handling department can analyze past claims data to select the optimal response method. For example, the claims handling department can select the optimal response method based on successful past claims handling cases. In this way, the optimal response method can be selected by referring to past claims data.

[0055] The claims handling unit can apply different response algorithms depending on the type of claim. For example, for claims regarding product defects, the claims handling unit can apply a rapid replacement algorithm. For example, for claims regarding service dissatisfaction, the claims handling unit can apply an algorithm that provides detailed explanations and improvement measures. For example, the claims handling unit can dynamically select the optimal response algorithm depending on the type of claim. This enables a quick and appropriate response by applying a response algorithm appropriate to the type of claim.

[0056] The claims handling department can adjust its response methods based on the reliability of the customer when handling claims. For example, the claims handling department can respond quickly and courteously to highly reliable customers, and cautiously to less reliable customers. The claims handling department can dynamically adjust its response methods according to the reliability of the customer. This enables quick and appropriate responses by considering the reliability of the customer.

[0057] The claims handling department can adjust its response methods by referring to relevant laws and regulations when handling claims. For example, the claims handling department sets standards for claim handling based on laws and regulations. For example, the claims handling department can dynamically adjust these standards in response to changes in laws and regulations. For example, the claims handling department can improve the accuracy of its responses by referring to relevant laws and regulations when handling claims. This enables a swift and appropriate response by referring to relevant laws and regulations.

[0058] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0059] The data extraction unit can improve accuracy by applying different extraction algorithms based on the invoice format. For example, if the invoice is handwritten, the data extraction unit can apply a handwriting recognition algorithm to extract the data. For example, if the invoice is in PDF format, the data extraction unit can apply a PDF analysis algorithm to extract the data. For example, if the invoice is in image format, the data extraction unit can apply an image recognition algorithm to extract the data. In this way, the accuracy of data extraction is improved by applying an extraction algorithm according to the invoice format.

[0060] The data extraction unit can determine the importance of data based on the content of the invoices and select data to extract preferentially. For example, the data extraction unit can prioritize extracting invoices with high invoice amounts. For example, the data extraction unit can prioritize extracting invoices with approaching payment deadlines. For example, the data extraction unit can prioritize extracting invoices from highly reliable trading partners. In this way, by determining the importance of data based on the content of the invoices, important data can be extracted preferentially.

[0061] The data extraction unit can apply different extraction methods based on the language and region of the invoice. For example, it can apply an English recognition algorithm to an English invoice, or a Japanese recognition algorithm to a Japanese invoice. For example, it can apply a region-specific extraction method to an invoice with a region-specific format. This improves the accuracy of data extraction by applying extraction methods tailored to the language and region of the invoice.

[0062] The data extraction unit can adjust the accuracy of data extraction based on the reliability of the invoice issuer. For example, the data extraction unit applies a high-precision extraction algorithm to invoices from reliable issuers. For example, the data extraction unit can carefully extract and verify data from invoices from unreliable issuers. For example, the data extraction unit can adjust the verification process of extracted data according to the reliability of the issuer. This improves the accuracy of data extraction by considering the reliability of the invoice issuer.

[0063] The contract information matching unit can adjust the matching timing based on the update frequency of contract information and transaction history. For example, if contract information is frequently updated, the contract information matching unit can shorten the matching timing. For example, if transaction history is frequently updated, the contract information matching unit can shorten the matching timing. For example, the contract information matching unit can dynamically adjust the matching timing according to the update frequency of contract information and transaction history. This improves the accuracy of matching by adjusting the matching timing based on the update frequency of contract information and transaction history.

[0064] The contract information matching unit can apply different matching algorithms based on the importance of the contract information during matching. For example, the contract information matching unit can apply a high-precision matching algorithm to contract information of high importance. For example, the contract information matching unit can apply a rapid matching algorithm to contract information of low importance. For example, the contract information matching unit can dynamically select a matching algorithm according to the importance of the contract information. This improves the accuracy of matching by applying a matching algorithm appropriate to the importance of the contract information.

[0065] The following briefly describes the processing flow for example form 1.

[0066] Step 1: The data extraction unit extracts data from the invoice. For example, it extracts invoice data using high-precision AIOCR and improves accuracy by applying different extraction algorithms depending on the invoice format. It can also determine the importance of the data based on the content of the invoice and select data to be extracted preferentially. Step 2: The contract information matching unit matches the data extracted by the data extraction unit against contract information and transaction history. For example, it can match invoice data with contract information and transaction history in real time and adjust the timing of the matching based on the update frequency of the contract information and transaction history. In addition, different matching algorithms can be applied depending on the importance of the contract information. Step 3: The anomaly detection unit detects anomalies based on the results verified by the contract information verification unit. For example, it can detect abnormal values ​​and fraudulent billing by comparing them with past data, and adjust the detection accuracy considering the reliability of the business partner. It can also adjust the detection criteria by referring to relevant laws and regulations. Step 4: The payment schedule optimization unit optimizes the payment schedule based on the anomalies detected by the anomaly detection unit. For example, it can create a payment schedule based on cash flow and customer priority, and dynamically adjust the payment schedule based on changes in cash flow. It can also optimize the payment schedule based on the credit information of customers. Step 5: The claims handling department handles claims based on the schedule optimized by the payment schedule optimization department. For example, it notifies relevant departments and business partners when a problem occurs and proposes a solution. It can select the optimal response method by referring to past claims data and apply different response algorithms depending on the type of claim.

[0067] (Example of form 2) The self-learning generation AI agent system, specialized for invoice processing and payment operations according to an embodiment of the present invention, is a system that extracts invoice data and enables real-time matching with contract information and transaction history. This system utilizes AIOCR to extract invoice data and enables real-time matching with contract information and transaction history. This provides a comprehensive solution for detecting inconsistencies, notifying anomalies, automatically creating optimal payment schedules, and handling claims quickly. Multiple AI agents cooperate to carry out tasks, immediately proposing solutions when problems occur, thereby streamlining the entire operation and improving reliability. For example, it starts with receiving invoices. For example, it collects invoices in paper or electronic format and handles various formats by utilizing the data classification capabilities of the generation AI agent. Next, it extracts invoice data (invoice amount, date, billing source, etc.) using high-precision AIOCR, and reduces misreadings and omissions by utilizing the learning capabilities of the generation AI agent. The extracted data is matched with contract information and transaction history in real time, and inconsistencies are detected. Rapid data integration is achieved by utilizing the knowledge graph of the generation AI agent. Furthermore, anomalies and fraudulent billing are detected by comparing with past data. Immediate identification of fraudulent billing is achieved by utilizing the anomaly detection algorithm of the generation AI agent. The system creates payment schedules considering cash flow and supplier priorities, and leverages the decision-making support capabilities of the generating AI agent to achieve optimal payment plans. If problems arise, it notifies relevant departments and suppliers and proposes solutions. Processing results are recorded to aid in future business improvements. The system utilizes the language generation capabilities of the generating AI agent to enable rapid and accurate complaint handling, and leverages its analytical capabilities to achieve continuous business improvement. In this way, it provides a comprehensive solution to streamline invoice processing and payment operations and improve reliability. This self-learning generating AI agent system, specialized for invoice processing and payment operations, can streamline and improve the reliability of these processes.

[0068] The self-learning generation AI agent system, specialized for invoice processing and payment operations according to the embodiment, comprises a data extraction unit, a contract information matching unit, an anomaly detection unit, a payment schedule optimization unit, and a claims handling unit. The data extraction unit extracts data from invoices. The data extraction unit extracts invoice data using, for example, high-precision AIOCR. The data extraction unit can improve accuracy by applying different extraction algorithms depending on the invoice format. The data extraction unit can determine importance based on the content of the invoice and select data to be extracted preferentially. The contract information matching unit matches the data extracted by the data extraction unit with contract information and transaction history. The contract information matching unit matches invoice data with contract information and transaction history in real time. The contract information matching unit can adjust the timing of matching based on the update frequency of contract information and transaction history. The contract information matching unit can apply different matching algorithms depending on the importance of the contract information. The anomaly detection unit detects anomalies based on the results matched by the contract information matching unit. The anomaly detection unit detects abnormal values ​​and fraudulent invoices by comparing them with past data. The anomaly detection unit can, for example, adjust the accuracy of its detection by considering the reliability of the trading partner when an anomaly is detected. The anomaly detection unit can, for example, adjust the detection criteria by referring to relevant laws and regulations when an anomaly is detected. The payment schedule optimization unit optimizes the payment schedule based on the anomalies detected by the anomaly detection unit. The payment schedule optimization unit creates a payment schedule based on, for example, cash flow and trading partner priority. The payment schedule optimization unit can dynamically adjust the payment schedule based on, for example, fluctuations in cash flow. The payment schedule optimization unit can optimize the payment schedule based on, for example, the credit information of the trading partner. The claims handling unit handles claims based on the schedule optimized by the payment schedule optimization unit. The claims handling unit notifies relevant departments and trading partners when a problem occurs and proposes a solution. The claims handling unit can, for example, select the optimal handling method by referring to past claims data when handling a claim.The claims handling unit can, for example, apply different response algorithms depending on the type of claim when handling a claim. This allows the self-learning, generation-type AI agent system, specialized for invoice processing and payment operations according to the embodiment, to streamline and improve the reliability of invoice processing and payment operations.

[0069] The data extraction unit extracts data from invoices. For example, the data extraction unit uses high-precision AIOCR to extract invoice data. Specifically, AIOCR is a technology that analyzes image data of invoices and converts it into text data. This allows for the accurate extraction of handwritten or printed invoice contents as digital data. The data extraction unit can improve accuracy by applying different extraction algorithms depending on the invoice format. Since invoice formats vary widely and differ from company to company, the data extraction unit selects the optimal algorithm based on pre-learned format information. For example, applying an algorithm specialized for a particular company's invoice format can improve extraction accuracy. The data extraction unit can also determine importance based on the invoice content and select data to be extracted preferentially. For example, it can prioritize the extraction of important data such as the invoice amount, payment deadline, and customer information, and manage it separately from other data. This allows the data extraction unit to quickly and accurately extract important information from invoices and utilize it for subsequent processing. Furthermore, the data extraction unit centrally manages the extracted data and can collaborate with other systems and departments as needed. For example, extracted data is stored on a cloud server, making it accessible to the contract information matching unit and the anomaly detection unit. Furthermore, by adjusting the frequency and accuracy of data extraction, flexible responses to specific situations and conditions become possible. This allows the data extraction unit to extract data efficiently and effectively, improving the overall system performance.

[0070] The contract information matching unit matches the data extracted by the data extraction unit with contract information and transaction history. For example, the contract information matching unit matches invoice data with contract information and transaction history in real time. Specifically, the contract information matching unit verifies whether the amount and transaction details stated on the invoice match the pre-registered contract information and past transaction history. This helps prevent fraudulent and erroneous billing. The contract information matching unit can adjust the timing of matching based on the update frequency of contract information and transaction history. For example, if contract information is frequently updated, the matching timing can be shortened to perform matching based on the latest information. Also, if there is a large transaction history, past data can be efficiently searched and matching can be performed quickly. The contract information matching unit can apply different matching algorithms depending on the importance of the contract information. For example, a more rigorous matching algorithm can be applied to high-value transactions and important contracts to improve accuracy. On the other hand, a simpler algorithm can be applied to relatively small transactions and general contracts to perform quick matching. In this way, the contract information matching unit can match data efficiently and accurately, improving the reliability of the entire system. Furthermore, the contract information verification unit can notify other departments and systems of the verification results in real time, supporting a rapid response. For example, if an anomaly is detected, it immediately notifies the anomaly detection unit, allowing appropriate countermeasures to be taken. In addition, based on the verification results, the payment schedule optimization unit provides information to create an optimal payment schedule. In this way, the contract information verification unit can improve the overall efficiency and reliability of the system.

[0071] The anomaly detection unit detects anomalies based on the results verified by the contract information verification unit. For example, the anomaly detection unit detects abnormal values ​​and fraudulent invoices by comparing them with past data. Specifically, the anomaly detection unit learns normal transaction patterns based on past transaction data and invoice data, and detects abnormal transactions based on that. For example, if there is an invoice that deviates significantly from the normal transaction amount or frequency, it is detected as an anomaly. For example, when detecting an anomaly, the anomaly detection unit can adjust the accuracy of detection by considering the reliability of the trading partner. For example, it can loosen the anomaly detection threshold for invoices from reliable trading partners and set a stricter threshold for invoices from unreliable trading partners. This improves the accuracy of anomaly detection. For example, when detecting an anomaly, the anomaly detection unit can adjust the detection criteria by referring to relevant laws and regulations. For example, it can set anomaly detection criteria based on laws and regulations in a specific industry or region and detect transactions that may violate those laws and regulations. This allows the anomaly detection unit to perform anomaly detection in compliance with laws and regulations. Furthermore, the anomaly detection unit provides information to respond quickly to detected anomalies. For example, if an anomaly is detected, the anomaly detection unit notifies the claims handling department of its detailed information to support appropriate action. Furthermore, the anomaly detection unit manages the history of detected anomalies, which can be used to improve the accuracy of future anomaly detection. This allows the anomaly detection unit to improve the overall reliability and safety of the system.

[0072] The payment schedule optimization unit optimizes the payment schedule based on anomalies detected by the anomaly detection unit. For example, the payment schedule optimization unit creates a payment schedule based on cash flow and supplier priorities. Specifically, the payment schedule optimization unit monitors the company's cash flow situation in real time and calculates the optimal payment timing. This optimizes the company's cash flow and reduces unnecessary spending. The payment schedule optimization unit can dynamically adjust the payment schedule based on fluctuations in cash flow. For example, it can adjust the payment schedule to coincide with periods when cash flow is prone to fluctuations, such as the end of the month or quarter, to stabilize cash flow. It can also optimize the payment schedule based on the credit information of suppliers. For example, it can make early payments to suppliers with high credit scores and delay payments to suppliers with low credit scores to mitigate risk. This allows the payment schedule optimization unit to optimize the company's cash flow and maintain good relationships with suppliers. Furthermore, the payment schedule optimization unit also responds appropriately to anomalies detected by the anomaly detection unit. For example, if fraudulent billing is detected, it will withhold payment and conduct a detailed investigation. Furthermore, once the anomaly is resolved, the payment schedule will be promptly readjusted and appropriate measures will be taken with the business partner. This allows the payment schedule optimization unit to respond flexibly and quickly to anomalies, thereby improving the company's reliability.

[0073] The Claims Handling Department handles claims based on a schedule optimized by the Payment Schedule Optimization Department. For example, when a problem occurs, the Claims Handling Department notifies relevant departments and business partners and proposes solutions. Specifically, when a problem occurs, the Claims Handling Department promptly notifies relevant departments and business partners and provides detailed information about the problem. This allows stakeholders to quickly begin taking action and resolve the problem early. For example, when handling a claim, the Claims Handling Department can select the optimal response method by referring to past claim data. For example, based on the response methods and results from similar problems in the past, it can select the optimal solution and resolve the problem quickly and effectively. Furthermore, the Claims Handling Department can apply different response algorithms depending on the type of claim. For example, for claims regarding payment delays, it readjusts the payment schedule, and for quality-related claims, it collaborates with the quality control department to implement countermeasures. This allows the Claims Handling Department to handle various types of claims appropriately and maintain trust with business partners. In addition, the Claims Handling Department can manage the history of claim handling and use it to improve the accuracy of future claim handling. For example, by analyzing past complaint handling history and identifying common problems and areas for improvement, the quality of future complaint handling can be enhanced. Furthermore, the complaint handling department can provide feedback on the results of complaint handling to relevant departments, which can be used to improve business processes. This allows the complaint handling department to improve the overall reliability and efficiency of the system.

[0074] The data extraction unit can extract invoice data using high-precision optical character recognition (OCR) technology. The data extraction unit can, for example, extract invoice data using high-precision optical character recognition technology. The data extraction unit can, for example, extract invoice data using an OCR engine. The data extraction unit can, for example, extract invoice data using techniques for improving accuracy. As a result, the accuracy of invoice data extraction is improved by using high-precision optical character recognition technology.

[0075] The contract information matching unit can match invoice data with contract information and transaction history in real time. For example, the contract information matching unit matches invoice data with contract information and transaction history in real time. The contract information matching unit can achieve real-time matching by, for example, increasing the matching frequency. The contract information matching unit can also achieve real-time matching by, for example, reducing the delay time. This enables rapid detection of discrepancies through real-time matching.

[0076] The anomaly detection unit can detect abnormal values ​​and fraudulent claims by comparing them with past data. For example, the anomaly detection unit can detect abnormal values ​​by comparing them with past data. For example, the anomaly detection unit can detect fraudulent claims by comparing them with past data. For example, the anomaly detection unit can detect abnormal values ​​by setting an abnormal value threshold. For example, the anomaly detection unit can detect fraudulent claims by analyzing patterns of fraudulent claims. This improves the accuracy of detecting abnormal values ​​and fraudulent claims by comparing them with past data.

[0077] The payment schedule optimization unit can create payment schedules based on cash flow and customer priority. For example, the payment schedule optimization unit can create payment schedules based on cash flow. For example, the payment schedule optimization unit can create payment schedules based on customer priority. For example, the payment schedule optimization unit can create payment schedules using optimization algorithms. For example, the payment schedule optimization unit can create payment schedules by setting the elements to be considered. This achieves an optimal payment schedule by considering cash flow and customer priority.

[0078] The complaints handling department can notify relevant departments and business partners when a problem occurs and propose solutions. For example, the complaints handling department can notify relevant departments when a problem occurs. For example, the complaints handling department can notify business partners when a problem occurs. For example, the complaints handling department can resolve problems by proposing solutions. For example, the complaints handling department can set up notification methods to notify relevant departments and business partners. This improves the reliability of operations by enabling a swift response to problems and the proposal of solutions.

[0079] The data extraction unit can estimate the user's emotions and adjust the timing of data extraction based on the estimated emotions. For example, if the user is stressed, the data extraction unit can delay the timing of data extraction. For example, the data extraction unit can perform data extraction when the user is relaxed. For example, if the user is in a hurry, the data extraction unit can speed up the timing of data extraction. For example, if the user is concentrating, the data extraction unit can adjust the timing of data extraction to avoid interrupting the user's work. In this way, the user's burden is reduced by adjusting the timing of data extraction according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0080] The data extraction unit can improve accuracy by applying different extraction algorithms based on the invoice format. For example, if the invoice is handwritten, the data extraction unit can apply a handwriting recognition algorithm to extract the data. For example, if the invoice is in PDF format, the data extraction unit can apply a PDF analysis algorithm to extract the data. For example, if the invoice is in image format, the data extraction unit can apply an image recognition algorithm to extract the data. In this way, the accuracy of data extraction is improved by applying an extraction algorithm according to the invoice format.

[0081] The data extraction unit can determine the importance of data based on the content of the invoices and select data to extract preferentially. For example, the data extraction unit can prioritize extracting invoices with high invoice amounts. For example, the data extraction unit can prioritize extracting invoices with approaching payment deadlines. For example, the data extraction unit can prioritize extracting invoices from highly reliable trading partners. In this way, by determining the importance of data based on the content of the invoices, important data can be extracted preferentially.

[0082] The data extraction unit can estimate the user's emotions and determine the priority of data to extract based on the estimated emotions. For example, if the user is stressed, the data extraction unit will postpone extracting less important data. For example, if the user is relaxed, the data extraction unit can prioritize extracting more important data. For example, if the user is in a hurry, the data extraction unit can prioritize extracting data that needs to be processed quickly. This reduces the user's burden by prioritizing data according to their emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may include, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0083] The data extraction unit can apply different extraction methods based on the language and region of the invoice. For example, it can apply an English recognition algorithm to an English invoice, or a Japanese recognition algorithm to a Japanese invoice. For example, it can apply a region-specific extraction method to an invoice with a region-specific format. This improves the accuracy of data extraction by applying extraction methods tailored to the language and region of the invoice.

[0084] The data extraction unit can adjust the accuracy of data extraction based on the reliability of the invoice issuer. For example, the data extraction unit applies a high-precision extraction algorithm to invoices from reliable issuers. For example, the data extraction unit can carefully extract and verify data from invoices from unreliable issuers. For example, the data extraction unit can adjust the verification process of extracted data according to the reliability of the issuer. This improves the accuracy of data extraction by considering the reliability of the invoice issuer.

[0085] The contract information matching unit can estimate the user's emotions and adjust the matching criteria based on the estimated emotions. For example, if the user is stressed, the contract information matching unit can loosen the matching criteria and process quickly. For example, if the user is relaxed, the contract information matching unit can tighten the matching criteria and process accurately. For example, if the user is in a hurry, the contract information matching unit can adjust the matching criteria to process quickly and accurately. This reduces the burden on the user by adjusting the matching criteria according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0086] The contract information matching unit can adjust the matching timing based on the update frequency of contract information and transaction history. For example, if contract information is frequently updated, the contract information matching unit can shorten the matching timing. For example, if transaction history is frequently updated, the contract information matching unit can shorten the matching timing. For example, the contract information matching unit can dynamically adjust the matching timing according to the update frequency of contract information and transaction history. This improves the accuracy of matching by adjusting the matching timing based on the update frequency of contract information and transaction history.

[0087] The contract information matching unit can apply different matching algorithms based on the importance of the contract information during matching. For example, the contract information matching unit can apply a high-precision matching algorithm to contract information of high importance. For example, the contract information matching unit can apply a rapid matching algorithm to contract information of low importance. For example, the contract information matching unit can dynamically select a matching algorithm according to the importance of the contract information. This improves the accuracy of matching by applying a matching algorithm appropriate to the importance of the contract information.

[0088] The contract information matching unit can estimate the user's emotions and adjust the display method of the matching results based on the estimated emotions. For example, if the user is tense, the contract information matching unit can provide a simple and highly visible display method. For example, if the user is relaxed, the contract information matching unit can provide a display method that includes detailed information. For example, if the user is in a hurry, the contract information matching unit can provide a display method that gets straight to the point. In this way, the burden on the user is reduced by adjusting the display method of the matching results according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0089] The contract information matching unit can apply different matching methods based on the region and language of the contract information and transaction history. For example, the contract information matching unit can apply an English recognition algorithm to English contract information. For example, the contract information matching unit can apply a Japanese recognition algorithm to Japanese contract information. For example, the contract information matching unit can apply a region-specific matching method to contract information with a region-specific format. This improves the accuracy of matching by applying matching methods according to the region and language of the contract information and transaction history.

[0090] The contract information matching unit can adjust the accuracy of the matching based on the reliability of the contract information. For example, the contract information matching unit applies a high-precision matching algorithm to highly reliable contract information. For example, the contract information matching unit can perform careful matching and verification on less reliable contract information. For example, the contract information matching unit can dynamically adjust the matching accuracy according to the reliability of the contract information. This improves the accuracy of the matching by considering the reliability of the contract information.

[0091] The anomaly detection unit can estimate the user's emotions and adjust the display method of the anomaly detection results based on the estimated user emotions. For example, if the user is tense, the anomaly detection unit can provide a simple and highly visible display method. For example, if the user is relaxed, the anomaly detection unit can provide a display method that includes detailed information. For example, if the user is in a hurry, the anomaly detection unit can provide a display method that gets straight to the point. In this way, the burden on the user is reduced by adjusting the display method of the anomaly detection results according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0092] The anomaly detection unit can adjust the detection accuracy based on the reliability of the trading partner when an anomaly is detected. For example, the anomaly detection unit applies a high-precision detection algorithm to data from highly reliable trading partners. For example, the anomaly detection unit can perform careful detection and verification on data from less reliable trading partners. For example, the anomaly detection unit can dynamically adjust the detection accuracy according to the reliability of the trading partner. This improves the accuracy of anomaly detection by considering the reliability of the trading partner.

[0093] The anomaly detection unit can adjust its detection criteria by referring to relevant laws and regulations when an anomaly is detected. For example, the anomaly detection unit sets the anomaly detection criteria based on laws and regulations. For example, the anomaly detection unit can dynamically adjust the anomaly detection criteria in response to changes in laws and regulations. For example, the anomaly detection unit can improve the accuracy of detection by referring to relevant laws and regulations when an anomaly is detected. This allows for appropriate adjustment of the anomaly detection criteria by referring to relevant laws and regulations.

[0094] The payment schedule optimization unit can estimate the user's emotions and determine the priority of the payment schedule based on the estimated emotions. For example, if the user is stressed, the payment schedule optimization unit will postpone less important payments. For example, if the user is relaxed, the payment schedule optimization unit can prioritize important payments. For example, if the user is in a hurry, the payment schedule optimization unit can prioritize payments that require quick processing. This reduces the user's burden by determining the priority of the payment schedule according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0095] The payment schedule optimization unit can dynamically adjust the payment schedule based on fluctuations in cash flow. For example, if cash flow deteriorates, the payment schedule optimization unit will review the payment schedule and postpone payments. For example, if cash flow improves, the payment schedule optimization unit can bring forward the payment schedule and make payments earlier. For example, the payment schedule optimization unit can dynamically adjust the payment schedule in response to fluctuations in cash flow to achieve an optimal payment plan. In this way, by adjusting the payment schedule in response to fluctuations in cash flow, an optimal payment plan is achieved.

[0096] The payment schedule optimization unit can adjust payment schedules based on the credit information of trading partners. For example, the payment schedule optimization unit can make payments early to trading partners with high credit scores. For example, the payment schedule optimization unit can delay payments to trading partners with low credit scores. For example, the payment schedule optimization unit can dynamically adjust payment schedules based on the credit information of trading partners to achieve an optimal payment plan. In this way, an optimal payment plan is achieved by optimizing payment schedules based on the credit information of trading partners.

[0097] The payment schedule optimization unit can estimate the user's emotions and adjust the display method of the payment schedule based on the estimated emotions. For example, if the user is stressed, the payment schedule optimization unit can provide a simple and highly visible display method. For example, if the user is relaxed, the payment schedule optimization unit can provide a display method that includes detailed information. For example, if the user is in a hurry, the payment schedule optimization unit can provide a display method that gets straight to the point. In this way, the burden on the user is reduced by adjusting the display method of the payment schedule according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0098] The payment schedule optimization unit can adjust the payment schedule based on the geographical distribution of trading partners when creating the payment schedule. For example, if trading partners are concentrated in a nearby area, the payment schedule optimization unit can make payments in a batch. For example, if trading partners are scattered far apart, the payment schedule optimization unit can make payments individually. For example, the payment schedule optimization unit can dynamically adjust the payment schedule based on the geographical distribution of trading partners to achieve an optimal payment plan. In this way, an optimal payment plan is achieved by taking into account the geographical distribution of trading partners.

[0099] The payment schedule optimization unit can optimize payment schedules by referring to relevant market data when creating them. For example, the payment schedule optimization unit adjusts payment schedules based on market trends. For example, the payment schedule optimization unit can adjust payment schedules based on market interest rate trends. For example, the payment schedule optimization unit can dynamically adjust payment schedules based on market supply and demand balance to achieve an optimal payment plan. This allows for the realization of an optimal payment plan by referring to relevant market data.

[0100] The complaint handling department can estimate the user's emotions and adjust its complaint handling method based on those emotions. For example, if the user is stressed, the complaint handling department can provide a quick and concise response. If the user is relaxed, the complaint handling department can provide a more detailed explanation. If the user is in a hurry, the complaint handling department can quickly offer a solution. This allows for a quick and appropriate response by adjusting the complaint handling method according to the user's emotions. Emotion estimation is achieved using emotion estimation functions, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0101] The claims handling department can select the optimal response method when handling a claim by referring to past claims data. For example, the claims handling department can refer to the response methods used in similar past claims. For example, the claims handling department can analyze past claims data to select the optimal response method. For example, the claims handling department can select the optimal response method based on successful past claims handling cases. In this way, the optimal response method can be selected by referring to past claims data.

[0102] The claims handling unit can apply different response algorithms depending on the type of claim. For example, for claims regarding product defects, the claims handling unit can apply a rapid replacement algorithm. For example, for claims regarding service dissatisfaction, the claims handling unit can apply an algorithm that provides detailed explanations and improvement measures. For example, the claims handling unit can dynamically select the optimal response algorithm depending on the type of claim. This enables a quick and appropriate response by applying a response algorithm appropriate to the type of claim.

[0103] The complaint handling department can estimate the user's emotions and determine the priority of complaint handling based on the estimated emotions. For example, if the user is stressed, the complaint handling department will respond quickly. For example, if the user is relaxed, the complaint handling department can provide a detailed response. For example, if the user is in a hurry, the complaint handling department can quickly offer a solution. This enables a quick and appropriate response by determining the priority of complaint handling according to the user's emotions. Emotion estimation is achieved using emotion estimation functions, such as emotion engines or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0104] The claims handling department can adjust its response methods based on the reliability of the customer when handling claims. For example, the claims handling department can respond quickly and courteously to highly reliable customers, and cautiously to less reliable customers. The claims handling department can dynamically adjust its response methods according to the reliability of the customer. This enables quick and appropriate responses by considering the reliability of the customer.

[0105] The claims handling department can adjust its response methods by referring to relevant laws and regulations when handling claims. For example, the claims handling department sets standards for claim handling based on laws and regulations. For example, the claims handling department can dynamically adjust these standards in response to changes in laws and regulations. For example, the claims handling department can improve the accuracy of its responses by referring to relevant laws and regulations when handling claims. This enables a swift and appropriate response by referring to relevant laws and regulations.

[0106] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0107] The data extraction unit can estimate the user's emotions and adjust the timing of data extraction based on the estimated emotions. For example, if the user is stressed, the data extraction unit can delay the timing of data extraction. For example, the data extraction unit can perform data extraction when the user is relaxed. For example, if the user is in a hurry, the data extraction unit can speed up the timing of data extraction. For example, if the user is concentrating, the data extraction unit can adjust the timing of data extraction to avoid interrupting the user's work. In this way, the user's burden is reduced by adjusting the timing of data extraction according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0108] The data extraction unit can improve accuracy by applying different extraction algorithms based on the invoice format. For example, if the invoice is handwritten, the data extraction unit can apply a handwriting recognition algorithm to extract the data. For example, if the invoice is in PDF format, the data extraction unit can apply a PDF analysis algorithm to extract the data. For example, if the invoice is in image format, the data extraction unit can apply an image recognition algorithm to extract the data. In this way, the accuracy of data extraction is improved by applying an extraction algorithm according to the invoice format.

[0109] The data extraction unit can determine the importance of data based on the content of the invoices and select data to extract preferentially. For example, the data extraction unit can prioritize extracting invoices with high invoice amounts. For example, the data extraction unit can prioritize extracting invoices with approaching payment deadlines. For example, the data extraction unit can prioritize extracting invoices from highly reliable trading partners. In this way, by determining the importance of data based on the content of the invoices, important data can be extracted preferentially.

[0110] The data extraction unit can estimate the user's emotions and determine the priority of data to extract based on the estimated emotions. For example, if the user is stressed, the data extraction unit will postpone extracting less important data. For example, if the user is relaxed, the data extraction unit can prioritize extracting more important data. For example, if the user is in a hurry, the data extraction unit can prioritize extracting data that needs to be processed quickly. This reduces the user's burden by prioritizing data according to their emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may include, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0111] The data extraction unit can apply different extraction methods based on the language and region of the invoice. For example, it can apply an English recognition algorithm to an English invoice, or a Japanese recognition algorithm to a Japanese invoice. For example, it can apply a region-specific extraction method to an invoice with a region-specific format. This improves the accuracy of data extraction by applying extraction methods tailored to the language and region of the invoice.

[0112] The data extraction unit can adjust the accuracy of data extraction based on the reliability of the invoice issuer. For example, the data extraction unit applies a high-precision extraction algorithm to invoices from reliable issuers. For example, the data extraction unit can carefully extract and verify data from invoices from unreliable issuers. For example, the data extraction unit can adjust the verification process of extracted data according to the reliability of the issuer. This improves the accuracy of data extraction by considering the reliability of the invoice issuer.

[0113] The contract information matching unit can estimate the user's emotions and adjust the matching criteria based on the estimated emotions. For example, if the user is stressed, the contract information matching unit can loosen the matching criteria and process quickly. For example, if the user is relaxed, the contract information matching unit can tighten the matching criteria and process accurately. For example, if the user is in a hurry, the contract information matching unit can adjust the matching criteria to process quickly and accurately. This reduces the burden on the user by adjusting the matching criteria according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0114] The contract information matching unit can adjust the matching timing based on the update frequency of contract information and transaction history. For example, if contract information is frequently updated, the contract information matching unit can shorten the matching timing. For example, if transaction history is frequently updated, the contract information matching unit can shorten the matching timing. For example, the contract information matching unit can dynamically adjust the matching timing according to the update frequency of contract information and transaction history. This improves the accuracy of matching by adjusting the matching timing based on the update frequency of contract information and transaction history.

[0115] The contract information matching unit can apply different matching algorithms based on the importance of the contract information during matching. For example, the contract information matching unit can apply a high-precision matching algorithm to contract information of high importance. For example, the contract information matching unit can apply a rapid matching algorithm to contract information of low importance. For example, the contract information matching unit can dynamically select a matching algorithm according to the importance of the contract information. This improves the accuracy of matching by applying a matching algorithm appropriate to the importance of the contract information.

[0116] The contract information matching unit can estimate the user's emotions and adjust the display method of the matching results based on the estimated emotions. For example, if the user is tense, the contract information matching unit can provide a simple and highly visible display method. For example, if the user is relaxed, the contract information matching unit can provide a display method that includes detailed information. For example, if the user is in a hurry, the contract information matching unit can provide a display method that gets straight to the point. In this way, the burden on the user is reduced by adjusting the display method of the matching results according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0117] The following briefly describes the processing flow for example form 2.

[0118] Step 1: The data extraction unit extracts data from the invoice. For example, it extracts invoice data using high-precision AIOCR and improves accuracy by applying different extraction algorithms depending on the invoice format. It can also determine the importance of the data based on the content of the invoice and select data to be extracted preferentially. Step 2: The contract information matching unit matches the data extracted by the data extraction unit against contract information and transaction history. For example, it can match invoice data with contract information and transaction history in real time and adjust the timing of the matching based on the update frequency of the contract information and transaction history. In addition, different matching algorithms can be applied depending on the importance of the contract information. Step 3: The anomaly detection unit detects anomalies based on the results verified by the contract information verification unit. For example, it can detect abnormal values ​​and fraudulent billing by comparing them with past data, and adjust the detection accuracy considering the reliability of the business partner. It can also adjust the detection criteria by referring to relevant laws and regulations. Step 4: The payment schedule optimization unit optimizes the payment schedule based on the anomalies detected by the anomaly detection unit. For example, it can create a payment schedule based on cash flow and customer priority, and dynamically adjust the payment schedule based on changes in cash flow. It can also optimize the payment schedule based on the credit information of customers. Step 5: The claims handling department handles claims based on the schedule optimized by the payment schedule optimization department. For example, it notifies relevant departments and business partners when a problem occurs and proposes a solution. It can select the optimal response method by referring to past claims data and apply different response algorithms depending on the type of claim.

[0119] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0120] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0121] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0122] Each of the multiple elements described above, including the data extraction unit, contract information matching unit, anomaly detection unit, payment schedule optimization unit, and claim handling unit, is implemented in at least one of the smart device 14 and the data processing device 12. For example, the data extraction unit extracts invoice data using the high-precision AIOCR of the smart device 14. The contract information matching unit is implemented in real time by the specific processing unit 290 of the data processing device 12, and matches invoice data with contract information and transaction history. The anomaly detection unit is implemented in real time by the specific processing unit 290 of the data processing device 12, and detects abnormal values ​​and fraudulent claims by comparing them with past data. The payment schedule optimization unit is implemented in real time by the specific processing unit 290 of the data processing device 12, and creates a payment schedule based on cash flow and customer priority. The claim handling unit is implemented in real time by the control unit 46A of the smart device 14, and notifies relevant departments and customers and proposes solutions when a problem occurs. The correspondence between each unit and the device or control unit is not limited to the example described above and can be changed in various ways.

[0123] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0124] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0125] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0126] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0127] The microphone 238 receives voice commands and other instructions from the user by receiving voice signals. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0128] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0129] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0130] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing by the processor 28. The storage 32 stores the specific processing program 56.

[0131] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0132] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0133] In the smart glasses 214, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0134] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0135] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0136] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0137] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0138] Each of the multiple elements described above, including the data extraction unit, contract information matching unit, anomaly detection unit, payment schedule optimization unit, and claim handling unit, is implemented by, for example, at least one of the smart glasses 214 and the data processing unit 12. For example, the data extraction unit extracts invoice data using the high-precision AIOCR of the smart glasses 214. The contract information matching unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and matches invoice data with contract information and transaction history in real time. The anomaly detection unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and detects abnormal values ​​and fraudulent claims by comparing them with past data. The payment schedule optimization unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and creates a payment schedule based on cash flow and customer priority. The claim handling unit is implemented by, for example, the control unit 46A of the smart glasses 214, and notifies relevant departments and customers and proposes solutions when a problem occurs. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0139] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0140] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0141] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0142] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0143] The microphone 238 receives voice commands and other instructions from the user by receiving voice signals. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0144] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0145] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0146] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0147] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0148] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0149] In the headset terminal 314, specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50 and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0150] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0151] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0152] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0153] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0154] Each of the multiple elements described above, including the data extraction unit, contract information matching unit, anomaly detection unit, payment schedule optimization unit, and claim handling unit, is implemented by, for example, at least one of the headset terminal 314 and the data processing unit 12. For example, the data extraction unit extracts invoice data using the high-precision AIOCR of the headset terminal 314. The contract information matching unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and matches invoice data with contract information and transaction history in real time. The anomaly detection unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and detects abnormal values ​​and fraudulent claims by comparing them with past data. The payment schedule optimization unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and creates a payment schedule based on cash flow and customer priority. The claim handling unit is implemented by, for example, the control unit 46A of the headset terminal 314, and notifies relevant departments and customers and proposes solutions when a problem occurs. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0155] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0156] As shown in Figure 7, the 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.

[0157] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0158] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0159] The microphone 238 receives voice commands and other instructions from the user by receiving voice signals. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0160] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0161] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0162] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0163] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0164] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0165] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0166] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0167] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0168] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0169] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0170] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0171] Each of the multiple elements described above, including the data extraction unit, contract information matching unit, anomaly detection unit, payment schedule optimization unit, and claim handling unit, is implemented by, for example, at least one of the robot 414 and the data processing unit 12. For example, the data extraction unit extracts invoice data using the high-precision AIOCR of the robot 414. The contract information matching unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and matches invoice data with contract information and transaction history in real time. The anomaly detection unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and detects abnormal values ​​and fraudulent claims by comparing them with past data. The payment schedule optimization unit is implemented by, for example, the specific processing unit 290 of the data processing unit 12, and creates a payment schedule based on cash flow and customer priority. The claim handling unit is implemented by, for example, the control unit 46A of the robot 414, and notifies relevant departments and customers and proposes solutions when a problem occurs. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0172] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0173] Figure 9 shows the emotion map 400, in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0174] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0175] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0176] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, and motorcycles, emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0177] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0178] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0179] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[0180] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0181] 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.

[0182] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0183] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0184] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0185] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0186] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0187] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0188] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and other things that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0189] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0190] (Note 1) A data extraction unit that extracts data from invoices, A contract information matching unit compares the data extracted by the data extraction unit with contract information and transaction history. An anomaly detection unit detects anomalies based on the results verified by the aforementioned contract information verification unit, A payment schedule optimization unit optimizes the payment schedule based on the anomaly detected by the anomaly detection unit, The payment schedule optimization unit includes a claims handling unit that handles claims based on a schedule optimized by the payment schedule optimization unit. A system characterized by the following features. (Note 2) The data extraction unit, Extract invoice data using high-precision optical character recognition technology. The system described in Appendix 1, characterized by the features described herein. (Note 3) The aforementioned contract information matching unit is: Match invoice data with contract information and transaction history in real time. The system described in Appendix 1, characterized by the features described herein. (Note 4) The aforementioned abnormality detection unit, Detect outliers and fraudulent claims by comparing them with historical data. The system described in Appendix 1, characterized by the features described herein. (Note 5) The aforementioned payment schedule optimization unit, Create a payment schedule based on cash flow and customer priorities. The system described in Appendix 1, characterized by the features described herein. (Note 6) The aforementioned claims handling unit is When a problem occurs, notify the relevant departments and business partners, and propose a solution. The system described in Appendix 1, characterized by the features described herein. (Note 7) The data extraction unit, We estimate the user's emotions and adjust the timing of data extraction based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 8) The data extraction unit, Apply different extraction algorithms based on the invoice format to improve accuracy. The system described in Appendix 1, characterized by the features described herein. (Note 9) The data extraction unit, The importance of each item is determined based on the invoice content, and data is selected to be extracted with priority. The system described in Appendix 1, characterized by the features described herein. (Note 10) The data extraction unit, It estimates the user's emotions and determines the priority of data to extract based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 11) The data extraction unit, Apply different extraction methods based on the language and region of the invoice. The system described in Appendix 1, characterized by the features described herein. (Note 12) The data extraction unit, Adjust the accuracy of data extraction based on the reliability of the invoice issuer. The system described in Appendix 1, characterized by the features described herein. (Note 13) The aforementioned contract information matching unit is: We estimate the user's sentiment and adjust the matching criteria based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 14) The aforementioned contract information matching unit is: Adjust the timing of reconciliation based on the update frequency of contract information and transaction history. The system described in Appendix 1, characterized by the features described herein. (Note 15) The aforementioned contract information matching unit is: During matching, different matching algorithms are applied based on the importance of the contract information. The system described in Appendix 1, characterized by the features described herein. (Note 16) The aforementioned contract information matching unit is: It estimates the user's emotions and adjusts how the matching results are displayed based on the estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 17) The aforementioned contract information matching unit is: Apply different matching methods based on the region and language of contract information and transaction history. The system described in Appendix 1, characterized by the features described herein. (Note 18) The aforementioned contract information matching unit is: Adjust the accuracy of the matching based on the reliability of the contract information. The system described in Appendix 1, characterized by the features described herein. (Note 19) The aforementioned abnormality detection unit, The system estimates the user's emotions and adjusts how anomaly detection results are displayed based on the estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 20) The aforementioned abnormality detection unit, When an anomaly is detected, the detection accuracy is adjusted based on the reliability of the business partner. The system described in Appendix 1, characterized by the features described herein. (Note 21) The aforementioned abnormality detection unit, When an anomaly is detected, the detection criteria are adjusted based on relevant laws and regulations. The system described in Appendix 1, characterized by the features described herein. (Note 22) The aforementioned payment schedule optimization unit, It estimates the user's emotions and prioritizes payment schedules based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 23) The aforementioned payment schedule optimization unit, Dynamically adjust payment schedules based on fluctuations in cash flow. The system described in Appendix 1, characterized by the features described herein. (Note 24) The aforementioned payment schedule optimization unit, Adjust payment schedules based on the credit information of business partners. The system described in Appendix 1, characterized by the features described herein. (Note 25) The aforementioned payment schedule optimization unit, It estimates the user's emotions and adjusts how the payment schedule is displayed based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 26) The aforementioned payment schedule optimization unit, When creating payment schedules, adjust them based on the geographical distribution of business partners. The system described in Appendix 1, characterized by the features described herein. (Note 27) The aforementioned payment schedule optimization unit, When creating payment schedules, adjust them based on relevant market data. The system described in Appendix 1, characterized by the features described herein. (Note 28) The aforementioned claims handling unit is We estimate the user's emotions and adjust our complaint handling methods based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 29) The aforementioned claims handling unit is When handling complaints, refer to past complaint data to select the most appropriate response method. The system described in Appendix 1, characterized by the features described herein. (Note 30) The aforementioned claims handling unit is When handling a complaint, different response algorithms are applied depending on the type of complaint. The system described in Appendix 1, characterized by the features described herein. (Note 31) The aforementioned claims handling unit is The system estimates the user's emotions and determines the priority of complaint handling based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 32) The aforementioned claims handling unit is When handling complaints, adjust the response method based on the reliability of the business partner. The system described in Appendix 1, characterized by the features described herein. (Note 33) The aforementioned claims handling unit is When handling a claim, adjust the response method based on the relevant laws and regulations. The system described in Appendix 1, characterized by the features described herein. [Explanation of Symbols]

[0191] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. A data extraction unit that extracts data from invoices, A contract information matching unit compares the data extracted by the data extraction unit with contract information and transaction history. An anomaly detection unit detects anomalies based on the results verified by the aforementioned contract information verification unit, A payment schedule optimization unit optimizes the payment schedule based on the anomaly detected by the anomaly detection unit, The payment schedule optimization unit includes a claims handling unit that handles claims based on a schedule optimized by the payment schedule optimization unit. A system characterized by the following features.

2. The data extraction unit, Extract invoice data using high-precision optical character recognition technology. The system according to feature 1.

3. The aforementioned contract information matching unit is: Match invoice data with contract information and transaction history in real time. The system according to feature 1.

4. The aforementioned abnormality detection unit, Detect outliers and fraudulent claims by comparing them with historical data. The system according to feature 1.

5. The aforementioned payment schedule optimization unit, Create a payment schedule based on cash flow and customer priorities. The system according to feature 1.

6. The aforementioned claims handling unit is When a problem occurs, notify the relevant departments and business partners, and propose a solution. The system according to feature 1.

7. The data extraction unit, We estimate the user's emotions and adjust the timing of data extraction based on the estimated user emotions. The system according to feature 1.

8. The data extraction unit, Apply different extraction algorithms based on the invoice format to improve accuracy. The system according to feature 1.

Images