Platform for building intelligent operation system based on enterprise management information
By introducing data acquisition and reconstruction, multi-dimensional modeling and knowledge graph technology into the enterprise management information system, the problems of management decision delay and decision basis deviation in the existing technology are solved, precise decision support and operation optimization are achieved, and enterprises are improved Operational efficiency and decision-making quality.
Patent Information
- Application Number
- CN202410958326.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-29
AI Technical Summary
The existing enterprise auxiliary decision-making system modeling products are based on massive data, massive calculations, and massive learning, resulting in inaccurate application scenarios, goals and results, and cannot solve problems such as delay in management decision-making, lack of data-driven insights, and deviations in management decision-making basis.
Provides a smart operation system platform built based on enterprise management information, including data acquisition and reconstruction module, data multi-dimensional modeling module, knowledge graph building module and information analysis and application module, through big data analysis engine, artificial intelligence technology, deep learning models and Knowledge graph technology, which carries out data collection, preprocessing, knowledge extraction, fusion and reasoning, realizes a structured and semantic-rich knowledge base, and supports applications such as business process optimization, risk management, decision support and customer relationship management.
It realizes the accuracy of application scenarios, goals and results, effectively solves problems such as delay in management decision making, lack of data-driven insights, and deviation in management decision-making basis. Through in-depth interactive analysis and generalization, it reduces the application difficulty of multi-dimensional computing and analysis services, and improves the application of Enterprise operation efficiency and decision-making quality.
Smart Images

Figure CN119046415A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of enterprise management platforms, and specifically to a platform for building an intelligent operation system based on enterprise management information. Background Art
[0002] Currently, the enterprise decision-making support system modeling products on the market are basically based on massive data, massive computing, and massive learning, which leads to inaccurate application scenarios, goals, and results, and cannot solve the current situation of delayed management decisions, lack of data-driven insights, and deviations in the basis for management decisions. The main problems are classified as follows:
[0003] Massive and complex data is difficult to process: First, there is an obvious lack of data collection and processing, which seems a bit too primitive for users who try to use the data directly, and the data is too redundant, which needs to be solved continuously through the "data access + data processing + data modeling" method; secondly, data processing skills are high, including data storage costs, data security risks, data maintenance costs, and data analysis difficulties. It is necessary to continuously decompose entity attributes to simplify storage and calculation; finally, there are high requirements for the performance of the basic layer of the data lake, and the data processing process must rely on high-performance servers. This is mainly due to problems caused by massive data, heterogeneous and diversified data, and delayed binding mode.
[0004] Limited by business logic: It is difficult to solidify complex business logic through ordinary algorithms. According to business logic, features that are helpful for decision-making evaluation are extracted and constructed, including operational data, personal records, behavioral data, etc. The quality of feature engineering directly affects the effect of the model. This involves big data technologies such as Hadoop, Spark, etc., as well as operations such as data cleaning, conversion and normalization.
[0005] Poor decision model construction and training effects: The decision model construction is scattered, and the relatively independent sub-models cannot support the decision demands of multiple data domains. After data access, the training results are difficult to achieve the expected goals. It is necessary to use machine learning algorithms to develop prediction models, such as logistic regression, decision trees, random forests, neural networks, etc. Optimize the model through cross-validation, A / B testing and other methods to ensure that the model has good predictive ability and interpretability.
[0006] Unprofessional decision engine construction: Currently, the enterprise decision-making tools in the industry all perform data retrieval and analysis on a single basic demand. The lack or incorrect construction of relevant decision engines will make it difficult to assist enterprise managers in making decisions. Technology that can integrate knowledge graphs can abstract management actions and scenarios for basic ERP management actions such as marketing, sales, procurement, and finance, develop decision engines, and convert models into actual decision logic, which can respond quickly and support real-time decision-making.
[0007] Therefore, those skilled in the art have provided a platform for building an intelligent operation system based on enterprise management information to solve the problems raised in the above background technology. Summary of the invention
[0008] The purpose of the present invention is to provide a platform for building an intelligent operation system based on enterprise management information, with precise application scenarios, goals and results, effectively solving the current situation of delayed management decision-making, lack of data-driven insights, and deviations in the basis for management decisions, so as to solve the problems raised in the above background technology.
[0009] To achieve the above object, the present invention provides the following technical solutions:
[0010] A platform for building a smart operation system based on enterprise management information, including a data collection and reconstruction module, a data multi-dimensional modeling module, a knowledge graph building module, and an information analysis and application module;
[0011] The data collection and reconstruction module is used to collect the required data, then transmit the collected data to the background data processing center for storage, and finally use the big data analysis engine and artificial intelligence technology to parse and reconstruct the stored data to extract valuable information; the data multidimensional modeling module is used to analyze, organize and simulate the company's business processes, operational data, and market information through data modeling technology; the knowledge graph construction module is used to start from comprehensive data collection, go through data preprocessing, knowledge extraction, knowledge fusion, and in-depth knowledge reasoning and analysis, and finally realize a structured and semantically rich knowledge base; the information analysis and application module is used to combine modeling and graphs to realize the application demands of business training models, predictions, and insight generation.
[0012] As a further solution of the present invention: the specific process of the knowledge graph building module to realize the knowledge base is:
[0013] Step 1: Data collection, obtaining data from data sources, including public datasets, professional databases, text materials and Web content;
[0014] Step 2: Data preprocessing: After collecting the data, clean and format the data;
[0015] Step 3: Knowledge extraction, extracting useful information and knowledge from the preprocessed data;
[0016] Step 4: Knowledge fusion, integrating knowledge from different data sources;
[0017] Step 5: Knowledge reasoning and analysis. Knowledge reasoning refers to the application of logical reasoning techniques in the knowledge graph to derive new knowledge or conclusions based on existing knowledge. Knowledge analysis refers to in-depth analysis of the knowledge graph to discover hidden relationships, patterns or trends between data.
[0018] As a further solution of the present invention: in the step 2, the data preprocessing also includes standardization processing, wherein the standardization processing includes unifying named entities and terms, and resolving inconsistencies in the data.
[0019] As a further solution of the present invention: the information analysis and application module can intelligently improve and optimize the control direction, specifically:
[0020] (1) Business process optimization: By analyzing the business processes of the enterprise, we can identify bottlenecks and problems in the processes, optimize the business processes, and improve the operational efficiency of the enterprise;
[0021] (2) Risk management: By modeling and analyzing enterprise operation data, potential risks can be predicted and data support can be provided for enterprise risk management;
[0022] (3) Decision support: Provide data-supported decision models for enterprise management to help enterprises make more reasonable decisions;
[0023] (4) Customer relationship management: By analyzing and modeling customer data, we can better understand customer needs and behaviors and improve customer satisfaction and loyalty;
[0024] (5) Human resource management: Model and analyze employee data to optimize human resource allocation and improve employee performance;
[0025] (6) Supply chain management: Optimize supply chain structure, reduce supply chain costs, and improve supply chain efficiency by modeling and analyzing supply chain data.
[0026] As a further solution of the present invention: the model constructed by the data multidimensional modeling module includes a statistical model, a machine learning model and a deep learning model.
[0027] As a further solution of the present invention: the data collected by the data collection and reconstruction module is stored in a distributed manner.
[0028] As a further solution of the present invention: the data acquisition and reconstruction module collects data in the following ways: sensor acquisition, Internet of Things device acquisition, and cloud data acquisition.
[0029] As a further solution of the present invention: the data acquisition and reconstruction module adopts real-time or periodic acquisition when performing data acquisition.
[0030] Compared with the prior art, the present invention has the following beneficial effects:
[0031] 1. The platform application scenarios, goals and results of this application are accurate, which can effectively solve the current situation of delayed management decision-making, lack of data-driven insights, and deviation in the basis of management decision-making.
[0032] 2. This application continuously innovates and optimizes by storing and processing large-scale and diverse data sets, from distributed storage to computing engines, from data governance to security protection.
[0033] 3. This application can be applied through the study of enterprise management graphs, combining the general big model with the big model application of business output characteristic enterprise scenarios.
[0034] 4. The data computing and reasoning technology of this application adopts deep graph algorithm and distributed graph algorithm to perform deep interactive analysis, one-click solidification and generalization, effectively reducing the application difficulty of multi-dimensional computing and analysis business.
[0035] 5. The knowledge graph technology of this application has a built-in knowledge extraction model for enterprise management scenarios, which can automatically extract events, entities, relationships, and attributes, and cooperate with the user's self-service training platform to achieve semi-automatic composition of unstructured data. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 A structural diagram of a platform for building an intelligent operation system based on enterprise management information. DETAILED DESCRIPTION
[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0038] As mentioned in the background technology of this application, research has found that the existing enterprise decision-making support system modeling products on the market are basically based on massive data, massive calculations, and massive learning, which leads to inaccurate application scenarios, goals, and results. They are unable to solve the current situation of delayed management decisions, lack of data-driven insights, and deviations in the basis for management decisions, and have certain defects.
[0039] In order to solve the above-mentioned defects, this application discloses a platform for building an intelligent operation system based on enterprise management information, with precise application scenarios, goals and results, which can effectively solve the current situation of delayed management decision-making, lack of data-driven insights, and deviations in the basis for management decisions.
[0040] The following will describe in detail how the solution of the present application solves the above technical problems in conjunction with the accompanying drawings.
[0041] See also Figure 1 In the embodiment of the present invention, a platform for building a smart operation system based on enterprise management information includes a data acquisition and reconstruction module, a data multidimensional modeling module, a knowledge graph construction module, and an information analysis and application module; the data acquisition and reconstruction module is used to collect the required data, and then transmit the collected data to the background data processing center for storage, and finally use the big data analysis engine and artificial intelligence technology to parse and reconstruct the stored data to extract valuable information; the data multidimensional modeling module is used to analyze, organize and simulate the business process, operation data, and market information of the enterprise through data modeling technology to improve enterprise management efficiency, optimize decision-making process and enhance enterprise core competitiveness; the knowledge graph construction module is used to start from comprehensive data collection, experience data preprocessing, knowledge extraction, knowledge fusion, and in-depth knowledge reasoning and analysis, and finally realize a structured and semantically rich knowledge base; the information analysis and application module is used to combine modeling and graphs to realize the training model of the business, predict and generate insights. The platform application scenarios, goals and results of this application are accurate, and effectively solve the current situation of management decision delays, lack of data-driven insights, and management decision-making basis deviations. It should be noted that in the data collection stage, the transmission speed, security and stability of the data need to be considered. In the data storage stage, it is necessary to consider the capacity, speed and reliability of data storage. Optimizing distributed storage computing technology can improve the performance and reliability of data storage. In the data analysis and reconstruction stage, it is necessary to consider the selection of data preprocessing, cleaning, integration and analysis methods. In addition, this application continues to innovate and optimize by storing and processing large-scale and diverse data sets, from distributed storage to computing engines, from data governance to security protection.
[0042] In this embodiment, the specific process of implementing the knowledge base by the knowledge graph building module is as follows: Step 1: Data collection, obtaining data from data sources, which include public data sets, professional databases, text materials and Web content. The quality and diversity of data are crucial to the comprehensiveness and accuracy of the knowledge graph; Step 2: Data preprocessing, after collecting the data, clean and format the data to remove noise and redundant information to ensure the quality of the data; Step 3: Knowledge extraction, extracting useful information and knowledge from the preprocessed data, among which entity recognition, relationship extraction and attribute extraction are the key tasks of knowledge extraction, which usually require the support of natural language processing technology and machine learning algorithms; Step 4: Knowledge fusion, integrating knowledge from different data sources, the goal of knowledge fusion is to solve the problems of entity redundancy, contradiction and inconsistency, and build a unified knowledge representation; Step 5: Knowledge reasoning and analysis, knowledge reasoning refers to the application of logical reasoning technology in the knowledge graph to derive new knowledge or conclusions based on existing knowledge, and knowledge analysis refers to in-depth analysis of the knowledge graph to discover hidden relationships, patterns or trends between data. Knowledge reasoning and analysis help to improve the value of the knowledge graph and support more complex decisions and insights. The knowledge graph technology of this application has a built-in knowledge extraction model for enterprise management scenarios, which can automatically extract events, entities, relationships, and attributes, and cooperate with the user's self-service training platform to achieve semi-automatic composition of unstructured data. It should be noted that entity recognition, relationship extraction, and attribute extraction are key technologies in natural language processing (NLP), and they play an important role in fields such as information extraction and knowledge graph construction. More specifically, entity recognition is an information extraction technology that aims to automatically identify entities with specific meanings from text data, such as names, place names, organization names, time expressions, numerical expressions, etc. These entities are basic elements in the text and are of great significance for understanding text content and building knowledge graphs. Relationship extraction is an information technology that extracts relationships between entities from natural language text. In the text, there are often various relationships between entities, such as "birthplace-character", "founder-company", etc. The goal of relationship extraction is to identify these relationships and structure them into a usable data format. Attribute extraction is an important research issue in the field of attribute-level sentiment analysis. Its purpose is to mine the attribute information of a specific entity from the text. The attribute information can be words or phrases that describe the characteristics of the entity, such as the function, appearance, price, etc. of the product. Natural language processing technology and machine learning algorithms are two important aspects of the field of artificial intelligence. They play a key role in processing and understanding human language. Among them, natural language processing technology is a technology that enables computers to understand, process and generate natural languages (such as Chinese, English, etc.). It covers tasks ranging from simple text processing to complex language understanding tasks such as speech recognition, machine translation, sentiment analysis, etc.A machine learning algorithm is an algorithm that enables computers to automatically learn from data. It does not require explicit programming rules to perform tasks, but instead discovers regularities and patterns by analyzing large amounts of data and makes predictions or decisions based on them.
[0043] In the present embodiment, in step two, data preprocessing also includes standardization processing, wherein the standardization processing includes unifying named entities and terms, and resolving inconsistencies in the data. Standardization processing of data preprocessing, also known as data standardization or normalization, is a commonly used data preprocessing step, which aims to convert data of different specifications or different distributions to a unified specification or a specific range to reduce the impact of scale, characteristics, distribution differences, etc. on the model. This operation is also called dimensionless because it eliminates the differences between different variables due to characteristic attributes such as properties, dimensions, and orders of magnitude, so that the values of each indicator are at the same order of magnitude, which is convenient for comprehensive analysis and comparison. The unified naming of entities and terms in this application is also a kind of standardization processing.
[0044] In this embodiment, the information analysis and application module can intelligently improve and optimize the management and control direction, specifically: (1) Business process optimization: By analyzing the business process of the enterprise, the bottlenecks and problems in the process are found, and then the business process is optimized to improve the operational efficiency of the enterprise. For example, the process mining technology is used to model the internal process of the enterprise, find out the redundant and inefficient links in the process, and provide improvement directions for the enterprise. Among them, process mining technology is a relatively new application of data mining in the field of workflow management. It aims to reproduce the real process of the business process by analyzing the logs generated by the operation of the workflow, and use this knowledge to analyze, monitor and improve the workflow; (2) Risk management: By modeling and analyzing the enterprise's operating data, potential risks are predicted, and data support is provided for enterprise risk management. For example, a risk assessment model is constructed, and the data on the enterprise's finance, market, credit, etc. are analyzed to assess the enterprise's risk level; (3) Decision support: Provide data-supported decision models for the enterprise management to help the enterprise Make more reasonable decisions, for example, build prediction models, analyze market trends, competitors and other information, and provide a basis for enterprise product pricing, market expansion and other decisions; (4) Customer relationship management: By analyzing and modeling customer data, better understand customer needs and behaviors, improve customer satisfaction and loyalty, for example, build a customer segmentation model, divide customers into different groups based on their consumption behavior, preferences and other information, and provide enterprises with targeted marketing strategies; (5) Human resource management: Model and analyze employee data, optimize human resource allocation, and improve employee performance. For example, build an employee performance prediction model, analyze employee work experience, skills and other factors, predict employee performance levels, and provide a basis for enterprise human resource management; (6) Supply chain management: By modeling and analyzing supply chain data, optimize supply chain structure, reduce supply chain costs, and improve supply chain efficiency. For example, build a supply chain optimization model, analyze data from suppliers, logistics, etc., and provide enterprises with the best supply chain strategy. Among them, enterprise information applications can support complex data scenarios, involving big data processing, machine learning, data mining and advanced data analysis. This application can combine the application of large models with business output characteristic enterprise scenarios based on the study of enterprise management maps. In addition, the data computing and reasoning technology of this application uses deep graph algorithms and distributed graph algorithms to perform deep interactive analysis, one-click solidification and generalization, effectively reducing the application difficulty of multi-dimensional computing and analysis services.
[0045] In this embodiment, the model constructed by the data multidimensional modeling module includes a statistical model, a machine learning model and a deep learning model. Statistical model, machine learning model and deep learning model are three important concepts in the field of data analysis and artificial intelligence, and they play a key role in data processing, prediction, classification and decision support. More specifically, a statistical model refers to a model established by mathematical statistical methods based on probability theory, which obtains the functional relationship between variables by analyzing the test or industrial process measurement data. A machine learning model is an algorithm that automatically learns and generates prediction results based on given input data (also called features). It uses a known data set to train the model and uses the model to predict or classify new and unknown data. A deep learning model is a machine learning method based on an artificial neural network (ANN). It constructs a network structure with multiple hidden layers by simulating the connection mode of human brain neurons, thereby realizing efficient feature learning and classification of complex data.
[0046] In this embodiment, the data collected by the data acquisition and reconstruction module is stored in a distributed manner. Distributed storage is a data storage technology that connects the disk space on each machine in the enterprise through a network, and forms these scattered storage resources into a virtual storage device, and the data is stored in various corners of the enterprise in a dispersed manner. More specifically, distributed storage is a technology that connects multiple storage devices through a network to jointly provide data storage and access services to the outside world. It uses multiple independent storage devices (nodes) to store data and reads and writes data through the network. Distributed storage has significant advantages in coping with large-scale data storage and management, and improving data reliability and security.
[0047] In this embodiment, the data acquisition and reconstruction module collects data in the following ways: sensor acquisition, IoT device acquisition, and cloud data acquisition. Sensors are the core devices for data acquisition, which can detect physical quantities and convert them into measurable and processable signals. IoT devices can realize information sharing and intercommunication between devices through built-in sensors and communication modules. These devices can automatically collect and transmit data, providing a rich data source for big data analysis. The data collected by IoT devices has the characteristics of strong real-time performance and high accuracy, and is widely used in smart homes, smart cities, industrial automation and other fields. Uploading device data to the cloud platform for storage and processing is another important data collection method. The cloud platform provides powerful data processing and analysis capabilities, which can realize real-time analysis and remote access of data. Through the cloud platform, users can view and analyze device data anytime and anywhere, providing strong support for decision-making.
[0048] In this embodiment, the data collection and reconstruction module uses real-time or periodic data collection. Real-time data collection and periodic data collection each have their own advantages and applicable scenarios. When choosing a data collection method, enterprises should make comprehensive considerations and trade-offs based on their own business needs and actual conditions.
[0049] The platform application scenarios, goals and results of the present invention are accurate, and effectively solve the current situation of delayed management decisions, lack of data-driven insights, and deviations in the basis of management decisions. The present application continuously innovates and optimizes by storing and processing large-scale and diverse data sets, from distributed storage to computing engines, from data governance to security protection. The present application can combine the large model application of business output characteristic enterprise scenarios on the basis of the study of enterprise management maps. The data calculation and reasoning technology of the present application adopts deep graph algorithms and distributed graph algorithms to perform deep interactive analysis, one-click solidification and generalization, effectively reducing the application difficulty of multi-dimensional computing and analysis services. The knowledge graph technology of the present application has a built-in knowledge extraction model for enterprise management scenarios, which can automatically extract events, entities, relationships, and attributes, and cooperate with the user's self-service training platform to achieve semi-automatic composition of unstructured data.
[0050] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
[0051] What is described above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A platform for building a smart operation system based on enterprise management information, characterized by: It includes data collection and reconstruction module, data multidimensional modeling module, knowledge graph building module and information analysis and application module; The data acquisition and reconstruction module is used to collect the required data, then transmit the collected data to the background data processing center for storage, and finally use the big data analysis engine and artificial intelligence technology to analyze and reconstruct the stored data to extract valuable information; The data multidimensional modeling module is used to analyze, organize and simulate the business processes, operational data and market information of the enterprise through data modeling technology; the knowledge graph building module is used to start from comprehensive data collection, go through data preprocessing, knowledge extraction, knowledge fusion, and in-depth knowledge reasoning and analysis, and finally realize a structured and semantically rich knowledge base; The information analysis and application module is used to combine modeling and graphing to realize the application demands of business training models, making predictions and generating insights.
2. The platform for building a smart operation system based on enterprise management information according to claim 1, characterized in that: The specific process of implementing the knowledge base by the knowledge graph building module is as follows: Step 1: Data collection, obtaining data from data sources, including public datasets, professional databases, text materials and Web content; Step 2: Data preprocessing: After collecting the data, clean and format the data; Step 3: Knowledge extraction, extracting useful information and knowledge from the preprocessed data; Step 4: Knowledge fusion, integrating knowledge from different data sources; Step 5: Knowledge reasoning and analysis. Knowledge reasoning refers to the application of logical reasoning techniques in the knowledge graph to derive new knowledge or conclusions based on existing knowledge. Knowledge analysis refers to in-depth analysis of the knowledge graph to discover hidden relationships, patterns or trends between data.
3. The platform for building a smart operation system based on enterprise management information according to claim 2 is characterized in that: In the step 2, data preprocessing also includes standardization processing, wherein the standardization processing includes unifying named entities and terms, and resolving inconsistencies in the data.
4. The platform for building a smart operation system based on enterprise management information according to claim 1, characterized in that: The information analysis and application module can intelligently improve and optimize the control direction, specifically: (1) Business process optimization: By analyzing the business processes of the enterprise, we can identify bottlenecks and problems in the processes, optimize the business processes, and improve the operational efficiency of the enterprise; (2) Risk management: By modeling and analyzing enterprise operation data, potential risks can be predicted and data support can be provided for enterprise risk management; (3) Decision support: Provide data-supported decision models for enterprise management to help enterprises make more reasonable decisions; (4) Customer relationship management: By analyzing and modeling customer data, we can better understand customer needs and behaviors and improve customer satisfaction and loyalty; (5) Human resource management: Model and analyze employee data to optimize human resource allocation and improve employee performance; (6) Supply chain management: Optimize supply chain structure, reduce supply chain costs, and improve supply chain efficiency by modeling and analyzing supply chain data.
5. The platform for building a smart operation system based on enterprise management information according to claim 1, characterized in that: The models constructed by the data multidimensional modeling module include statistical models, machine learning models and deep learning models.
6. The platform for building a smart operation system based on enterprise management information according to claim 1, characterized in that: The data collected by the data collection and reconstruction module is stored in a distributed manner.
7. The platform for building a smart operation system based on enterprise management information according to claim 1, characterized in that: The data acquisition and reconstruction module collects data in the following ways: sensor acquisition, IoT device acquisition, and cloud data acquisition.
8. The platform for building a smart operation system based on enterprise management information according to claim 1, characterized in that: The data acquisition and reconstruction module collects data in real time or periodically.
Citation Information
Patent Citations
Operation and maintenance management system and method based on knowledge base
CN114706994A
Intelligent enterprise decision-making method based on large model
CN116976564A
Assistant decision-making system based on knowledge enhancement large model under background of group-type enterprise
CN118069716A
Risk prediction method and system for business operations
WO2021103492A1
AIOT paas internet of things operation platform
WO2022257181A1
Cited By
Full-dimensional digital operation decision-making method and system and storage medium
CN119378826A
Multi-scene multi-base large model engine system
CN120011187A
