Intelligent operation management system based on big data analysis

Through an intelligent operation management system based on big data analysis, we can identify and predict enterprise risks and generate early warning signals, and solve the problems of inefficient analysis and inaccurate risk prediction in the existing technology, and achieve more accurate and efficient operation management.

CN119026925AInactive Publication Date: 2024-11-26MSC(DALIAN)INFO TECH CO LID

Patent Information

Application Number
CN202411513799.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-29
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In the prior art, the analysis of operation management methods is inefficient and the risk prediction is inaccurate, making it difficult for enterprises to identify and respond to potential risks in a timely manner.

Method used

Provide an intelligent operation management system based on big data analysis. By collecting enterprise monitoring data blocks, identifying and extracting key risk factors, using risk prediction algorithms to generate risk warning indexes, and output corresponding warning signals. The system includes an operation management unit, a risk factor mining unit, an associated extraction unit, a risk prediction unit and a risk warning unit. Through the coordinated work of multiple units, risk prediction and early warning can be achieved.

Benefits of technology

Improve operational efficiency, accurately predict the future risk situation of the company, reduce the time and human resource consumption of manual analysis, and reduce the inaccuracy of risk prediction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026925A_ABST
    Figure CN119026925A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent operation management system based on big data analysis, and relates to the related field of data processing, and the system comprises an operation management unit which is used for receiving an enterprise monitoring data block of a target enterprise; the risk factor mining unit is used for carrying out risk factor mining to obtain an enterprise risk factor set; the risk factor extraction unit is used for extracting a first enterprise risk factor; the association extraction unit is used for performing association data extraction on the enterprise monitoring data block to obtain a first monitoring association data block; the risk prediction unit is used for carrying out risk prediction and outputting a first risk early warning index; the risk early warning unit is used for generating a first enterprise risk early warning signal. The technical problems that an existing operation management method is low in analysis efficiency and inaccurate in risk prediction are solved, and the technical effects of improving the operation efficiency and accurately predicting the future risk condition of an enterprise are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to data processing related fields, and in particular to an intelligent operation management system based on big data analysis. Background Art

[0002] With the rapid development of information technology, big data has become an important resource for enterprise operation and management. In a highly competitive environment, enterprises need to understand their own operating conditions in real time and accurately in order to identify and respond to potential risks in a timely manner. In existing methods, enterprises usually evaluate their own operating conditions and risk levels through regular financial reports, market surveys or internal audits. Although these methods can provide certain reference information, due to the large and complex amount of data, manual analysis requires a lot of time and human resources, and is easily affected by subjective factors, resulting in low analysis efficiency and inaccurate risk prediction.

[0003] Among the current related technologies, the operational management methods have technical problems such as low analysis efficiency and inaccurate risk prediction. Summary of the invention

[0004] This application achieves the technical effect of improving operational efficiency and accurately predicting the future risk situation of the enterprise by providing an intelligent operation management system based on big data analysis, collecting enterprise monitoring data blocks, identifying and extracting key risk factors, analyzing based on enterprise risk prediction algorithms, outputting risk warning indexes, and generating corresponding warning signals.

[0005] This application provides an intelligent operation management system based on big data analysis, including: An operation management unit, the operation management unit is used to receive an enterprise monitoring data block of a target enterprise; a risk factor mining unit, the risk factor mining unit is used to perform risk factor mining on the target enterprise to obtain an enterprise risk factor set, wherein the enterprise risk factor set is a collection of multiple risk factors, and each risk factor represents a risk faced by the enterprise in operation; a risk factor extraction unit, the risk factor extraction unit is used to extract a first enterprise risk factor based on the enterprise risk factor set, wherein the first enterprise risk factor includes any one of the enterprise risk factors in the enterprise risk factor set; an association extraction unit, the association extraction unit is used to perform association data extraction on the enterprise monitoring data block based on the first enterprise risk factor to obtain a first monitoring association data block; a risk prediction unit, the risk prediction unit is used to perform risk prediction on the target enterprise based on an embedded enterprise risk prediction algorithm in combination with the first monitoring association data block, and output a first risk warning index, the enterprise The risk prediction algorithm includes a risk prediction learning unit, a data input unit and a first risk warning index generation unit, wherein the risk prediction learning unit is used to perform risk prediction learning according to the first enterprise risk factor and construct a first enterprise risk prediction component, wherein the first enterprise risk prediction component includes a first window enterprise risk prediction channel and a second window enterprise risk prediction channel, the data input unit is used to input the first monitoring-related data block into the first window enterprise risk prediction channel to obtain the first window enterprise risk prediction index, input the first monitoring-related data block into the second window enterprise risk prediction channel, and output the second window enterprise risk prediction index, the first risk warning index generation unit is used to input the first window enterprise risk prediction index and the second window enterprise risk prediction index into a risk warning analysis function to generate the first risk warning index; a risk warning unit, wherein the risk warning unit is used to generate a first enterprise risk warning signal based on the first risk warning index.

[0006] In a possible implementation, the risk factor mining unit includes: A mining enterprise source setting unit, the mining enterprise source setting unit is used to set a mining enterprise source according to the target enterprise, wherein the mining enterprise source includes the target enterprise and multiple enterprises of the same type as the target enterprise; a risk event retrieval unit, the risk event retrieval unit is used to perform risk event retrieval according to the mining enterprise source to obtain an enterprise risk event set; a risk indicator identification unit, the risk indicator identification unit is used to perform risk indicator identification according to the enterprise risk event set to obtain a risk indicator identification set; a duplicate indicator cleaning unit, the duplicate indicator cleaning unit is used to perform duplicate indicator cleaning according to the risk indicator identification set to obtain multiple sample enterprise risk indicators; a similar feature cleaning unit, the similar feature cleaning unit is used to perform similar feature cleaning according to the multiple sample enterprise risk indicators to generate the enterprise risk factor set.

[0007] In a possible implementation, the similar feature cleaning unit includes: A pairwise similarity feature evaluation unit, the pairwise similarity feature evaluation unit is used to perform pairwise similarity feature evaluation based on the multiple sample enterprise risk indicators to obtain multiple risk indicator similarity indexes; a classification unit, the classification unit is used to classify the multiple sample enterprise risk indicators based on the multiple risk indicator similarity indexes to obtain a first enterprise risk factor set less than a risk indicator similarity threshold, and a first enterprise risk indicator set greater than / equal to the risk indicator similarity threshold; a similarity indicator fusion unit, the similarity indicator fusion unit is used to perform similarity indicator fusion based on the first enterprise risk indicator set to obtain a second enterprise risk factor set; a set adding unit, the set adding unit is used to add the first enterprise risk factor set and the second enterprise risk factor set to the enterprise risk factor set.

[0008] In a possible implementation, the first risk warning index generating unit includes: A risk warning analysis function construction unit, wherein the risk warning analysis function construction unit is used for the risk warning analysis function: ; Among them, RAI represents the first risk warning index, ERFW represents the first predetermined gain coefficient, ERF represents the first window enterprise risk prediction index, ERFO represents the first window enterprise risk prediction expectation index, ERCW represents the second predetermined gain coefficient, ERFW and ERCW are both positive numbers, and the sum of ERFW and ERCW is 1, ERC represents the second window enterprise risk prediction index, and ERCO represents the second window enterprise risk prediction expectation index.

[0009] In a possible implementation, the risk prediction learning unit includes: A window activation unit, the window activation unit is used to activate a first enterprise risk prediction time zone window and a second enterprise risk prediction time zone window, wherein the first enterprise risk prediction time zone window is larger than the second enterprise risk prediction time zone window; a risk prediction retrieval learning unit, the risk prediction retrieval learning unit is used to perform risk prediction retrieval learning according to the first enterprise risk factor and the first enterprise risk prediction time zone window, and build the first window enterprise risk prediction channel, perform risk prediction retrieval learning according to the first enterprise risk factor and the second enterprise risk prediction time zone window, and build the second window enterprise risk prediction channel; a first enterprise risk prediction component establishment unit, the first enterprise risk prediction component establishment unit is used to establish the first enterprise risk prediction component according to the first window enterprise risk prediction channel and the second window enterprise risk prediction channel.

[0010] In a possible implementation, the risk prediction retrieval learning unit includes: An enterprise risk record collection unit, the enterprise risk record collection unit is used to collect enterprise risk records according to the first enterprise risk factor and the first enterprise risk prediction time zone window, and obtain multiple enterprise risk record groups, wherein each enterprise risk record group includes a historical monitoring associated data block corresponding to the first enterprise risk factor, and a first window enterprise risk history index corresponding to the first enterprise risk prediction time zone window; a risk prediction learner activation unit, the risk prediction learner activation unit is used to activate the risk prediction learner; a supervised training unit, the supervised training unit is used to supervise the risk prediction learner according to the multiple enterprise risk record groups to obtain a risk prediction error coefficient; a first window enterprise risk prediction channel generation unit, the first window enterprise risk prediction channel generation unit is used to generate the first window enterprise risk prediction channel when the risk prediction error coefficient is less than a risk prediction error threshold.

[0011] In a possible implementation, the risk warning unit includes: A risk warning construction unit, the risk warning construction unit is used for the risk warning unit including a risk warning verification component and an enterprise risk grading warning component; a risk warning index input unit, the risk warning index input unit is used to input the first risk warning index into the risk warning verification component; a judgment unit, the judgment unit is used to judge whether the first risk warning index is greater than / equal to a preset risk warning index in the risk warning verification component; an enterprise risk grading warning component activation unit, the enterprise risk grading warning component activation unit is used to activate the enterprise risk grading warning component if the first risk warning index is greater than / equal to the preset risk warning index; an enterprise risk first warning signal generation unit, the enterprise risk first warning signal generation unit is used to input the first risk warning index into the enterprise risk grading warning component to generate the enterprise risk first warning signal.

[0012] It is planned to adopt the intelligent operation management system based on big data analysis proposed in this application, receive the enterprise monitoring data block of the target enterprise through the operation management unit, conduct risk factor mining on the target enterprise through the risk factor mining unit to obtain the enterprise risk factor set, extract the first enterprise risk factor according to the enterprise risk factor set through the risk factor extraction unit, wherein the first enterprise risk factor includes any enterprise risk factor in the enterprise risk factor set, perform associated data extraction on the enterprise monitoring data block based on the first enterprise risk factor through the associated extraction unit to obtain the first monitoring associated data block, conduct risk prediction on the target enterprise based on the embedded enterprise risk prediction algorithm through the risk prediction unit in combination with the first monitoring associated data block, output the first risk warning index, and generate the first enterprise risk warning signal based on the first risk warning index through the risk warning unit, thereby achieving the technical effect of improving operation efficiency and accurately predicting the future risk situation of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. It should be understood that the previous or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0014] Figure 1 A schematic diagram of the structure of an intelligent operation management system based on big data analysis provided in an embodiment of the present application.

[0015] Figure 2 A schematic diagram of the structure of a risk factor mining unit of an intelligent operation management system based on big data analysis provided in an embodiment of the present application.

[0016] Explanation of reference numerals: operation management unit 10 , risk factor mining unit 20 , risk factor extraction unit 30 , correlation extraction unit 40 , risk prediction unit 50 , risk warning unit 60 . DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The present application embodiment provides an intelligent operation management system based on big data analysis, such as Figure 1 As shown, the system comprises: The operation management unit 10 is used to receive the enterprise monitoring data block of the target enterprise. Specifically, the operation management unit 10 continuously monitors and receives the monitoring data block from the target enterprise through various methods such as API interface, data pipeline, file transfer, etc., that is, various data collected from the target enterprise, including various operation information of the enterprise, including financial data, production data, sales data, supply chain data, etc. The received data can be structured (such as table data in the database) or unstructured (such as text data, pictures, videos, etc. on social media). After receiving the data, the operation management unit 10 performs basic verification on the data to ensure the integrity, accuracy and compliance of the data. If problems are found in the data (such as format errors, missing values, abnormal values, etc.), records are made and corresponding processing processes (such as data cleaning, data completion, etc.) are triggered. The verified data is stored in the data warehouse or database of the system for subsequent analysis and mining. The operation management unit 10 synchronizes with the data source of the target enterprise regularly or in real time. During the synchronization process, if new data or data updates are found, corresponding processing is performed and updated to the data warehouse of the system.

[0021] The risk factor mining unit 20 is used to mine risk factors for the target enterprise and obtain an enterprise risk factor set. The enterprise risk factor set is a set containing multiple risk factors, and each risk factor represents a risk faced by the enterprise in operation. Specifically, the risk factor mining unit 20 obtains the enterprise monitoring data block of the target enterprise from the operation management unit 10, and performs appropriate preprocessing on the data according to the characteristics of the data and the mining target, such as data cleaning, data conversion, data aggregation, etc. The preprocessed data is analyzed in depth using data mining techniques (such as association analysis, cluster analysis, text mining, etc.) and machine learning algorithms (such as decision trees, neural networks, support vector machines, etc.) to identify patterns, trends or anomalies hidden in the data that may be related to enterprise risks. The identified risk factors are quantitatively or qualitatively evaluated to determine their potential impact on the enterprise operation. According to the evaluation results, the risk factors are sorted or classified for subsequent risk management and response. The identified and evaluated risk factors are sorted into an enterprise risk factor set, which is a set containing multiple risk factors, and each risk factor represents a risk that may be faced in the enterprise operation.

[0022] like Figure 2As shown, in a possible implementation, the risk factor mining unit 20 includes: a mining enterprise source setting unit, the mining enterprise source setting unit is used to set the mining enterprise source according to the target enterprise, wherein the mining enterprise source includes the target enterprise and multiple enterprises of the same type as the target enterprise. Specifically, the target enterprise is the starting point for risk factor mining. According to the characteristics of the target enterprise such as the industry and scale, multiple enterprises of the same type are selected as references. These enterprises of the same type have similarities with the target enterprise in terms of business model, market environment, etc. The target enterprise and the selected enterprises of the same type are set as mining enterprise sources, that is, data source enterprises for risk factor mining. A risk event retrieval unit, the risk event retrieval unit is used to perform risk event retrieval according to the mining enterprise source to obtain an enterprise risk event set. Specifically, historical risk events are retrieved from the mining enterprise source, including financial crisis, legal proceedings, supply chain interruption, etc., and the retrieved risk events are sorted and analyzed to form a comprehensive enterprise risk event set. A risk indicator identification unit, the risk indicator identification unit is used to identify risk indicators according to the enterprise risk event set to obtain a risk indicator identification set. Specifically, the enterprise risk event set is deeply analyzed to identify indicators or factors associated with the occurrence of risk events, including financial indicators (such as debt ratio, current ratio, etc.), market indicators (such as market share, number of competitors, etc.) or operational indicators (such as inventory turnover rate, employee turnover rate, etc.), and the identified risk indicators are organized into a risk indicator identification set. A duplicate indicator cleaning unit is used to clean duplicate indicators according to the risk indicator identification set to obtain multiple sample enterprise risk indicators. Specifically, the risk indicator identification set is cleaned to remove duplicate or highly correlated indicators. After cleaning, multiple independent and representative sample enterprise risk indicators are obtained. A similar feature cleaning unit is used to clean similar features according to the multiple sample enterprise risk indicators to generate the enterprise risk factor set. Specifically, the multiple sample enterprise risk indicators after cleaning are further checked to remove those indicators that are different but have similar features or behaviors in essence. For example, some financial indicators may be obtained through different calculation formulas, but they actually reflect the same financial status of the enterprise. Therefore, it is necessary to remove redundant indicators therein and finally generate a streamlined and efficient enterprise risk factor set. This implementation method builds a systematic and structured risk factor mining process. Each unit has its unique function and purpose, working together to identify the risk factors that the target enterprise may face. Through a gradually in-depth analysis and cleaning process, it achieves the technical effect of ensuring that the final set of enterprise risk factors is both comprehensive and accurate.

[0023] In a possible implementation, the similarity feature cleaning unit includes: a pairwise similarity feature evaluation unit, which is used to perform pairwise similarity feature evaluation based on the multiple sample enterprise risk indicators to obtain multiple risk indicator similarity indexes. Specifically, multiple sample enterprise risk indicators are used as input data, and each risk indicator is compared pairwise, and the similarity between them is calculated by methods such as cosine similarity and Pearson correlation coefficient, and the calculated similarity value is converted into a risk indicator similarity index. The risk indicator similarity index reflects the similarity between different risk indicators and is a quantitative indicator for measuring the similarity between two risk indicators. A classification unit, which is used to classify the multiple sample enterprise risk indicators based on the multiple risk indicator similarity indexes, and obtain a first enterprise risk factor set less than a risk indicator similarity threshold, and a first enterprise risk indicator set greater than / equal to the risk indicator similarity threshold. Specifically, a risk indicator similarity threshold is pre-set to determine which risk indicators are considered similar, and according to the risk indicator similarity index, multiple sample enterprise risk indicators are divided into two categories: a first enterprise risk factor set less than the risk indicator similarity threshold, and a first enterprise risk indicator set greater than or equal to the risk indicator similarity threshold. A similar indicator fusion unit, the similar indicator fusion unit is used to fuse similar indicators according to the first enterprise risk indicator set to obtain a second enterprise risk factor set. Specifically, those risk indicators with high similarity are selected from the first enterprise risk indicator set, and these risk indicators with high similarity are fused to reduce redundancy and generate more representative risk indicators. The fusion method includes taking the average value, the maximum value, the minimum value, etc., and the fused risk indicators are output as the second enterprise risk factor set. A set adding unit, the set adding unit is used to add the first enterprise risk factor set and the second enterprise risk factor set to the enterprise risk factor set. Specifically, the first enterprise risk factor set (risk factors less than the risk indicator similarity threshold) and the second enterprise risk factor set (fused risk factors) are merged, and the merged risk factor set is added to the enterprise risk factor set. This implementation method constructs a more systematic and structured risk factor mining process. By identifying those similar or redundant risk indicators, it avoids the use of repeated or highly correlated indicators in subsequent analysis, improves the accuracy of analysis, and achieves the technical effect of improving the accuracy and efficiency of risk factor mining.

[0024] The risk factor extraction unit 30 is used to extract a first enterprise risk factor according to the enterprise risk factor set, wherein the first enterprise risk factor includes any one enterprise risk factor in the enterprise risk factor set. Specifically, the risk factor extraction unit 30 receives the enterprise risk factor set from the risk factor mining unit 20. According to a preset rule, algorithm or user's instruction, a risk factor selection strategy (based on multiple factors such as the historical importance of the risk factor, current environmental conditions, and specific trends in the industry) is determined, and according to the selection strategy, a risk factor is extracted from the enterprise risk factor set as the first enterprise risk factor.

[0025] The associated extraction unit 40 is used to extract associated data from the enterprise monitoring data block based on the first enterprise risk factor to obtain a first monitoring associated data block. Specifically, the associated extraction unit 40 receives the first enterprise risk factor from the risk factor extraction unit 30 and receives the enterprise monitoring data block from the operation management unit 10. The associated extraction unit 40 performs in-depth analysis on the enterprise monitoring data block and uses data analysis techniques (such as association rule mining, time series analysis, text mining, etc.) to identify data fields, data patterns or data trends related to the first enterprise risk factor. According to the analysis results, data directly or indirectly related to the first enterprise risk factor is extracted from the enterprise monitoring data block, including financial indicators, production data, sales data, supply chain information, customer feedback, etc. associated with the risk factor. The extracted associated data is integrated to form a unified data structure or data table, and the data is cleaned to remove duplicate, erroneous or irrelevant data to ensure the accuracy and validity of the data. The integrated and cleaned associated data is packaged into a first monitoring associated data block, which contains all key information closely related to the first enterprise risk factor for subsequent risk prediction and early warning.

[0026] The risk prediction unit 50 is used to predict the risk of the target enterprise based on the embedded enterprise risk prediction algorithm and in combination with the first monitoring association data block, and output the first risk warning index. The enterprise risk prediction algorithm includes a risk prediction learning unit, a data input unit and a first risk warning index generation unit. The risk prediction learning unit is used to perform risk prediction learning according to the first enterprise risk factor and construct a first enterprise risk prediction component, wherein the first enterprise risk prediction component includes a first window enterprise risk prediction channel and a second window enterprise risk prediction channel. The data input unit is used to input the first monitoring association data block into the first window enterprise risk prediction channel to obtain the first window enterprise risk prediction index, input the first monitoring association data block into the second window enterprise risk prediction channel, and output the second window enterprise risk prediction index. The first risk warning index generation unit is used to input the first window enterprise risk prediction index and the second window enterprise risk prediction index into the risk warning analysis function to generate the first risk warning index. Specifically, the risk prediction unit 50 receives the first monitoring association data block from the association extraction unit 40, and performs necessary preprocessing on the received first monitoring association data block, such as data standardization, missing value filling, abnormal value processing, etc., to ensure the quality and consistency of the data. The data input unit inputs the first monitoring associated data block into the first window enterprise risk prediction channel and the second window enterprise risk prediction channel respectively. These two channels are constructed by the risk prediction learning unit according to the first enterprise risk factor and different enterprise risk prediction time zone windows. In the first window enterprise risk prediction channel, the data is processed to generate the first window enterprise risk prediction index, which represents the risk prediction result based on a wider time zone window (the first enterprise risk prediction time zone window). In the second window enterprise risk prediction channel, the data is processed to generate the second window enterprise risk prediction index, which represents the risk prediction result based on a narrower time zone window (the second enterprise risk prediction time zone window). The first risk warning index generation unit uses the risk warning parsing function to combine the first window enterprise risk prediction index, the second window enterprise risk prediction index and their respective expected indexes to calculate the first risk warning index, which is a composite indicator that integrates the prediction results of different time zone windows and reflects the overall risk status of the target enterprise in the current and future period.

[0027] In a possible implementation, the first risk warning index generating unit includes: a risk warning analysis function building unit, and the risk warning analysis function building unit is used for the risk warning analysis function to be: ; Among them, RAI represents the first risk warning index, ERFW represents the first predetermined gain coefficient, ERF represents the first window enterprise risk prediction index, ERFO represents the first window enterprise risk prediction expected index, ERCW represents the second predetermined gain coefficient, ERFW and ERCW are both positive numbers, and the sum of ERFW and ERCW is 1, ERC represents the second window enterprise risk prediction index, and ERCO represents the second window enterprise risk prediction expected index. This implementation method introduces a risk warning analysis function construction unit, which is used to construct a risk warning analysis function to generate the first risk warning index by combining the first window enterprise risk prediction index and the second window enterprise risk prediction index. The risk warning analysis function achieves the technical effect of generating a comprehensive and accurate first risk warning index through flexible weight allocation and standardized calculation methods.

[0028] In a possible implementation, the risk prediction learning unit includes: a window activation unit, the window activation unit is used to activate the first enterprise risk prediction time zone window and the second enterprise risk prediction time zone window, wherein the first enterprise risk prediction time zone window is larger than the second enterprise risk prediction time zone window. Specifically, according to the preset rules or strategies, the duration of the first enterprise risk prediction time zone window (long-term window) and the second enterprise risk prediction time zone window (short-term window) is determined. For example, the long-term window can be set to the data of the past year, and the short-term window can be set to the data of the past month. In the risk prediction process, according to the current time point and the time period required for prediction, the corresponding time zone window is activated. A risk prediction retrieval learning unit, the risk prediction retrieval learning unit is used to perform risk prediction retrieval learning according to the first enterprise risk factor and the first enterprise risk prediction time zone window, build the first window enterprise risk prediction channel, perform risk prediction retrieval learning according to the first enterprise risk factor and the second enterprise risk prediction time zone window, and build the second window enterprise risk prediction channel. Specifically, according to the activated time zone window, historical data related to the first enterprise risk factor is retrieved from the data warehouse or data source, the retrieved data is cleaned, converted and formatted to meet the requirements of the enterprise risk prediction model, and key features are extracted from the preprocessed data for building the enterprise risk prediction model. The enterprise risk prediction model is trained using the extracted features and corresponding labels (such as historical risk events). For the first window enterprise risk prediction channel, the data in the long-term window is used for training; for the second window enterprise risk prediction channel, the data in the short-term window is used for training. The first enterprise risk prediction component establishment unit is used to establish the first enterprise risk prediction component according to the first window enterprise risk prediction channel and the second window enterprise risk prediction channel. Specifically, the prediction results of the first window enterprise risk prediction channel and the second window enterprise risk prediction channel are integrated through weighted average, voting or other integration strategies, and the integrated prediction results are output in a visual form or in the form of a data interface. This implementation method realizes a comprehensive and accurate enterprise risk prediction system. By activating different time zone windows, it captures risk changes at different time scales and provides rich prediction information. By integrating the prediction results at different time scales, a comprehensive and accurate prediction result is obtained, achieving the technical effect of improving the accuracy and reliability of risk prediction.

[0029] In a possible implementation, the risk prediction retrieval learning unit includes: an enterprise risk record collection unit, the enterprise risk record collection unit is used to collect enterprise risk records according to the first enterprise risk factor and the first enterprise risk prediction time zone window, and obtain multiple enterprise risk record groups, wherein each enterprise risk record group includes a historical monitoring associated data block corresponding to the first enterprise risk factor, and a first window enterprise risk history index corresponding to the first enterprise risk prediction time zone window. Specifically, according to the first enterprise risk factor and the first enterprise risk prediction time zone window, determine the data range to be collected, retrieve the historical monitoring data related to the first enterprise risk factor from the historical database of the target enterprise, and ensure that the time range of these data is within the first enterprise risk prediction time zone window, group the collected data according to the risk factor or time sequence, and form multiple enterprise risk record groups, each enterprise risk record group includes a historical monitoring associated data block corresponding to the first enterprise risk factor (such as sales data, financial data, etc.), and a first window enterprise risk history index corresponding to the first enterprise risk prediction time zone window (which can be a historical risk score or rating). A risk prediction learner activation unit, the risk prediction learner activation unit is used to activate the risk prediction learner. Specifically, according to the needs, a suitable machine learning model (such as BP neural network, residual neural network, etc.) is selected as a risk prediction learner (a machine learning model for learning and predicting enterprise risks), the risk prediction learner is activated to receive training data, and necessary parameters (such as learning rate, number of iterations, etc.) are set. A supervised training unit, the supervised training unit is used to supervise the risk prediction learner according to the multiple enterprise risk record groups to obtain a risk prediction error coefficient. Specifically, multiple enterprise risk record groups are used as training data and input into the risk prediction learner, the risk prediction learner is trained using the training data, the parameters of the model are adjusted to minimize the prediction error, and during or after the training, the prediction error of the model on the validation set is calculated and saved as the risk prediction error coefficient. A first window enterprise risk prediction channel generation unit, the first window enterprise risk prediction channel generation unit is used to generate the first window enterprise risk prediction channel when the risk prediction error coefficient is less than the risk prediction error threshold. Specifically, it is checked whether the risk prediction error coefficient is less than the preset risk prediction error threshold. If the risk prediction error coefficient meets the requirement, the model training is considered successful, and the first window enterprise risk prediction channel is generated. The first window enterprise risk prediction channel is a machine learning model that has been trained and can be used to predict the enterprise risk in the first enterprise risk prediction time zone window. This implementation method generates an enterprise risk prediction channel that can be used for actual prediction by training the machine learning model, achieving the technical effect of improving the accuracy and efficiency of risk prediction.

[0030] The risk warning unit 60 is used to generate a first enterprise risk warning signal based on the first risk warning index. Specifically, the risk warning unit 60 receives the first risk warning index from the risk prediction unit 50, and judges the first risk warning index according to the preset warning rules and thresholds, wherein the preset warning rules and thresholds are set based on historical data, industry standards or specific needs of the enterprise. If the first risk warning index exceeds the set threshold, the risk warning unit 60 generates a first enterprise risk warning signal, wherein the enterprise risk first warning signal can be classified according to the severity and urgency of the risk, for example, divided into three levels of low, medium and high, and different warning signal levels correspond to different risk response measures and priorities. The generated first enterprise risk warning signal is output to the system user or decision maker, and the output form may include but is not limited to e-mail, SMS, system notification, report, etc., to ensure the timely communication and reception of information. The embodiment of the present application adopts technical means such as collecting enterprise monitoring data blocks, identifying and extracting key risk factors, analyzing based on enterprise risk prediction algorithms, outputting risk warning indexes, and generating corresponding warning signals, so as to achieve the technical effect of improving operational efficiency and accurately predicting the future risk situation of the enterprise.

[0031] In a possible implementation, the risk warning unit 60 includes: a risk warning construction unit, the risk warning construction unit is used for the risk warning unit 60 to include a risk warning verification component and an enterprise risk classification warning component. Specifically, the risk warning verification component and the enterprise risk classification warning component are set and initialized, a preset risk warning index (i.e., risk threshold) is configured for the risk warning verification component, and warning rules of different risk levels are configured for the enterprise risk classification warning component. Among them, the risk warning verification component is a component used to verify whether the first risk warning index reaches the preset threshold, and the enterprise risk classification warning component is a component that divides the risk into different levels and generates corresponding warning signals according to the size of the first risk warning index. A risk warning index input unit, the risk warning index input unit is used to input the first risk warning index into the risk warning verification component; a judgment unit, the judgment unit is used to judge whether the first risk warning index is greater than / equal to the preset risk warning index in the risk warning verification component. Specifically, the first risk warning index is compared with the preset risk warning index in the risk warning verification component. If the first risk warning index is greater than or equal to the preset risk warning index, a positive result is output; otherwise, a negative result is output. An enterprise risk grading warning component activation unit, the enterprise risk grading warning component activation unit is used to activate the enterprise risk grading warning component if the first risk warning index is greater than / equal to the preset risk warning index. Specifically, according to the output result of the judgment unit, it is judged whether the enterprise risk grading warning component needs to be activated. If the judgment result is affirmative, the enterprise risk grading warning component is activated to prepare for risk grading warning. An enterprise risk first warning signal generation unit, the enterprise risk first warning signal generation unit is used to input the first risk warning index into the enterprise risk grading warning component to generate the enterprise risk first warning signal. Specifically, the first risk warning index is input into the activated enterprise risk grading warning component, and the first risk warning index is risk graded according to the preset rules of the enterprise risk grading warning component, and the corresponding enterprise risk first warning signal is generated according to the grading result. This implementation method ensures that only when the first risk warning index reaches or exceeds the preset risk warning index, the subsequent warning operation will be triggered, avoiding unnecessary waste of resources and achieving the technical effect of improving the efficiency and accuracy of the target enterprise risk management.

[0032] Although the present application makes various references to certain units in the system according to the embodiments of the present application, any number of different units may be used and run on the user terminal and / or server, and the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0033] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. Intelligent operation management system based on big data analysis, characterized by: The system comprises: An operation management unit, the operation management unit is used to receive an enterprise monitoring data block of a target enterprise; A risk factor mining unit, wherein the risk factor mining unit is used to mine risk factors for the target enterprise to obtain an enterprise risk factor set, wherein the enterprise risk factor set is a set including multiple risk factors, each risk factor representing a risk faced by the enterprise in operation; a risk factor extraction unit, the risk factor extraction unit being used to extract a first enterprise risk factor according to the enterprise risk factor set, wherein the first enterprise risk factor includes any one enterprise risk factor in the enterprise risk factor set; A correlation extraction unit, the correlation extraction unit is used to extract correlation data from the enterprise monitoring data block based on the first enterprise risk factor to obtain a first monitoring correlation data block; A risk prediction unit, wherein the risk prediction unit is used to perform risk prediction on the target enterprise based on an embedded enterprise risk prediction algorithm in combination with the first monitoring-related data block, and output a first risk warning index. The enterprise risk prediction algorithm includes a risk prediction learning unit, a data input unit, and a first risk warning index generation unit. The risk prediction learning unit is used to perform risk prediction learning according to the first enterprise risk factor and construct a first enterprise risk prediction component, wherein the first enterprise risk prediction component includes a first window enterprise risk prediction channel and a second window enterprise risk prediction channel. The data input unit is used to input the first monitoring-related data block into the first window enterprise risk prediction channel to obtain a first window enterprise risk prediction index, input the first monitoring-related data block into the second window enterprise risk prediction channel, and output a second window enterprise risk prediction index. The first risk warning index generation unit is used to input the first window enterprise risk prediction index and the second window enterprise risk prediction index into a risk warning analysis function to generate the first risk warning index. A risk warning unit is used to generate a first warning signal of enterprise risk based on the first risk warning index.

2. The intelligent operation management system based on big data analysis according to claim 1, characterized in that: The risk factor mining unit includes: A mining enterprise source setting unit, the mining enterprise source setting unit is used to set a mining enterprise source according to the target enterprise, wherein the mining enterprise source includes the target enterprise and multiple enterprises of the same type as the target enterprise; A risk event retrieval unit, the risk event retrieval unit is used to perform risk event retrieval according to the mined enterprise source to obtain an enterprise risk event set; A risk indicator identification unit, the risk indicator identification unit is used to identify risk indicators according to the enterprise risk event set to obtain a risk indicator identification set; A duplicate indicator cleaning unit, the duplicate indicator cleaning unit is used to clean duplicate indicators according to the risk indicator identification set to obtain multiple sample enterprise risk indicators; A similar feature cleaning unit is used to perform similar feature cleaning according to the multiple sample enterprise risk indicators to generate the enterprise risk factor set.

3. The intelligent operation management system based on big data analysis according to claim 2, characterized in that: The similar feature cleaning unit comprises: A pairwise similarity feature evaluation unit, the pairwise similarity feature evaluation unit is used to perform pairwise similarity feature evaluation based on the multiple sample enterprise risk indicators to obtain multiple risk indicator similarity indexes; A classification unit, the classification unit is used to classify the multiple sample enterprise risk indicators based on the multiple risk indicator similarity indexes, and obtain a first enterprise risk factor set that is less than a risk indicator similarity threshold, and a first enterprise risk indicator set that is greater than / equal to the risk indicator similarity threshold; A similarity indicator fusion unit, the similarity indicator fusion unit is used to fuse similarity indicators according to the first enterprise risk indicator set to obtain a second enterprise risk factor set; A set adding unit, the set adding unit is used to add the first enterprise risk factor set and the second enterprise risk factor set to the enterprise risk factor set.

4. The intelligent operation management system based on big data analysis according to claim 1, characterized in that: The first risk warning index generating unit comprises: A risk warning analysis function construction unit, wherein the risk warning analysis function construction unit is used for the risk warning analysis function: ; Among them, RAI represents the first risk warning index, ERFW represents the first predetermined gain coefficient, ERF represents the first window enterprise risk prediction index, ERFO represents the first window enterprise risk prediction expectation index, ERCW represents the second predetermined gain coefficient, ERFW and ERCW are both positive numbers, and the sum of ERFW and ERCW is 1, ERC represents the second window enterprise risk prediction index, and ERCO represents the second window enterprise risk prediction expectation index.

5. The intelligent operation management system based on big data analysis according to claim 1, characterized in that: The risk prediction learning unit includes: A window activation unit, the window activation unit is used to activate a first enterprise risk prediction time zone window and a second enterprise risk prediction time zone window, wherein the first enterprise risk prediction time zone window is larger than the second enterprise risk prediction time zone window; A risk prediction retrieval learning unit, the risk prediction retrieval learning unit is used to perform risk prediction retrieval learning according to the first enterprise risk factor and the first enterprise risk prediction time zone window, and build the first window enterprise risk prediction channel; perform risk prediction retrieval learning according to the first enterprise risk factor and the second enterprise risk prediction time zone window, and build the second window enterprise risk prediction channel; The first enterprise risk prediction component establishing unit is used to establish the first enterprise risk prediction component according to the first window enterprise risk prediction channel and the second window enterprise risk prediction channel.

6. The intelligent operation management system based on big data analysis according to claim 5, characterized in that: The risk prediction retrieval learning unit includes: An enterprise risk record collection unit, the enterprise risk record collection unit is used to collect enterprise risk records according to the first enterprise risk factor and the first enterprise risk prediction time zone window, and obtain multiple enterprise risk record groups, wherein each enterprise risk record group includes a historical monitoring associated data block corresponding to the first enterprise risk factor, and a first window enterprise risk history index corresponding to the first enterprise risk prediction time zone window; A risk prediction learner activation unit, wherein the risk prediction learner activation unit is used to activate the risk prediction learner; A supervised training unit, the supervised training unit is used to perform supervised training on the risk prediction learner according to the multiple enterprise risk record groups to obtain a risk prediction error coefficient; The first window enterprise risk prediction channel generation unit is used to generate the first window enterprise risk prediction channel when the risk prediction error coefficient is less than the risk prediction error threshold.

7. The intelligent operation management system based on big data analysis according to claim 1, characterized in that: The risk early warning unit comprises: A risk warning construction unit, wherein the risk warning construction unit is used for the risk warning unit to include a risk warning verification component and an enterprise risk classification warning component; A risk warning index input unit, the risk warning index input unit is used to input the first risk warning index into the risk warning verification component; A judgment unit, the judgment unit is used to judge whether the first risk warning index is greater than / equal to a preset risk warning index in the risk warning verification component; An enterprise risk classification warning component activation unit, the enterprise risk classification warning component activation unit is used to activate the enterprise risk classification warning component if the first risk warning index is greater than / equal to the preset risk warning index; An enterprise risk first warning signal generating unit is used to input the first risk warning index into the enterprise risk classification warning component to generate the enterprise risk first warning signal.

Citation Information

Patent Citations

  • Energy power operation evaluation index screening method and device

    CN111985815A

  • Enterprise risk early warning method and device

    CN114066242A

  • Sub-enterprise identification method and device, storage medium and electronic equipment

    CN115496364A

  • Enterprise risk prediction method and device, storage medium, electronic equipment and product

    CN115759751A

  • Risk identification system based on enterprise management

    CN117788187A

Cited By

  • Enterprise risk assessment method based on multi-agent risk assessment algorithm MA-ERC

    CN120822841A

  • Enterprise risk assessment method based on multi-agent risk assessment algorithm MA-ERC

    CN120822841B