Enterprise strategic intelligent early warning method, device, electronic equipment and storage medium

Abstract

The present invention relates to the field of artificial intelligence and provides an enterprise strategic intelligent early warning method. The method performs NLP recognition processing on external operating data to extract external feature data; performs standardization processing and coding processing on the enterprise's internal evaluation data to obtain internal reference data; obtains primary reference information through a multiple regression model, obtains auxiliary reference information through a temporal differential autoregressive sliding average model, obtains a predicted customer throughput trend based on the primary and auxiliary reference information, and obtains current enterprise risk warning information based on the customer throughput trend and preset trend risk comparison information. In this way, intelligent early warning of customer risks is achieved, and possible losses discovered afterwards are reduced. In other words, intelligent early warning of customer risks is achieved, and possible losses discovered afterwards are reduced. The accuracy and practicality of the model are evaluated to improve the throughput prediction accuracy.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, to data analysis technology, and in particular to an enterprise strategic intelligent early warning method, device, electronic device and computer-readable storage medium. Background Art

[0002] Due to historical limitations and technological constraints, most port groups currently rely on customer visits as their primary means of acquiring key customer information. However, irregular visit cycles and incomplete information collection severely impact key customer risk management. In most cases, key customer risk information is only received after the fact, resulting in a lack of awareness of customer situations, delayed responses, and a failure to provide early or timely warnings. Once risks occur, port companies can suffer significant losses. Therefore, proactively identifying customer business risk behaviors and formulating timely strategies are crucial to the healthy development of port companies.

[0003] Port companies' primary business objective is passenger and cargo throughput, and achieving this throughput depends on the business scale and operational performance of their customers (i.e., cargo owners). Generally speaking, the larger the scale and concentration of a customer within a company, the greater the likelihood of dependence on that customer, and customer risk can easily translate into operational risk for the company. Key customer risks for port companies primarily include two aspects: customer volume risk and customer churn risk.

[0004] Therefore, there is an urgent need for a strategic intelligent early warning method for port enterprises based on data mining that integrates customer business volume risk and customer churn risk to achieve intelligent early warning of customer risks and reduce possible losses discovered afterwards. Summary of the Invention

[0005] The present invention provides an enterprise strategic intelligent early warning method to solve the problem in the existing technology that due to factors such as limitations in historical business development experience and technical limitations, the main way for most port groups to obtain information on major customers is customer visits. However, factors such as irregular visit cycles and incomplete information dimensions seriously affect the risk control of major customers. In most cases, the risk information of major customers is received after the fact, resulting in unclear perception of customer conditions, delayed responses, and failure to provide early or timely warnings. Once a risk occurs, it is easy to cause significant losses to port enterprises.

[0006] To achieve the above objectives, the present invention provides an enterprise strategic intelligent early warning method, comprising:

[0007] Based on a preset database or data retrieval platform, external business data is retrieved based on fields related to pre-established external business environment themes, and internal enterprise assessment data is retrieved based on fields related to pre-established internal assessment themes;

[0008] Performing natural language recognition processing on the external business data to extract external feature data; performing standardization and coding processing on the internal enterprise evaluation data to obtain internal reference data;

[0009] Performing node averaging processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal differential autoregressive sliding average model; constructing a regression model using the internal reference data as factors to obtain a multivariate regression model;

[0010] The primary reference information is obtained through the multivariate regression model, the auxiliary reference information is obtained through the temporal differential autoregressive sliding average model, the customer throughput trend is predicted based on the primary reference information and the auxiliary reference information, and the risk warning information of the current enterprise is obtained based on the customer throughput trend and preset trend risk control information.

[0011] Optionally, the method of retrieving external business data based on a preset external business environment subject field and retrieving enterprise internal assessment data based on a preset internal assessment subject field based on a preset database or data retrieval platform includes:

[0012] Divide the internal and external data of the preset business into external data sets and internal data sets; wherein the external data set includes at least macroeconomic data, industry price and industry development data, and media data; the internal data set includes customer business volume data, enterprise basic information data, and enterprise operation data within the enterprise group;

[0013] Establishing an external business environment theme and an internal enterprise evaluation theme for the preset business, and marking the external business environment theme on the external data set and the internal enterprise evaluation theme on the internal data set;

[0014] The external business data is crawled around the fields of the external business environment theme, and the internal evaluation data of the enterprise is retrieved around the fields of the internal evaluation theme; wherein, the external business data at least includes macroeconomic environment analysis, policy impact analysis and industry technology development public opinion impact analysis.

[0015] Optionally, the retrieving internal enterprise evaluation data includes:

[0016] Obtain enterprise asset size changes, business scope changes, management changes, administrative penalty labels, and public opinion labels through pre-set enterprise assessment plug-ins;

[0017] Assign values ​​to the changes in the enterprise's asset scale, business scope, management level, administrative penalty label, and public opinion label to obtain an internal assessment data table for the enterprise;

[0018] The enterprise internal evaluation data table is traversed and calculated using a preset evaluation algorithm to obtain the enterprise internal evaluation data.

[0019] Optionally, obtaining the macroeconomic environment analysis includes:

[0020] The customer's throughput in the enterprise is obtained through the preset information crawling plug-in, and the current GDP price, average exchange rate, and import and export policy labels are obtained through the preset media information plug-in;

[0021] Creating a table to be filled in according to a preset arrangement rule, and mapping the throughput, the current price of GDP, the average exchange rate, and the import and export policy label in the table to be filled in to form a macroeconomic environment analysis table;

[0022] The macroeconomic environment analysis table is input into a preset intelligent information extraction model, so that the intelligent information extraction model automatically outputs a macroeconomic environment analysis according to the macroeconomic environment analysis table.

[0023] Optionally, performing natural language recognition processing on the external business data to extract external feature data includes:

[0024] Performing algorithm fitting based on pre-acquired sample data about a preset business to obtain an NLP recognition model for starting the NLP semantic recognition service;

[0025] Based on the NLP recognition model, training is performed using the sample data related to the preset business to obtain an NLP semantic recognition service process;

[0026] The external business data is processed to form standard data, and the standard data is input into the NLP semantic recognition service process, so that the NLP semantic recognition service process performs semantic recognition on the external business data to obtain enterprise-related keywords and data corresponding to the keywords; wherein the process of processing the external business data to form standard data is to change the external business data into a format in which one item corresponds to one data;

[0027] Synonym replacement is performed on the keywords to obtain near-meaning keywords, and the keywords, data corresponding to the keywords, and the near-meaning keywords are packaged to form external feature data.

[0028] Optionally, the standardizing and coding the enterprise internal evaluation data to obtain internal reference data includes:

[0029] Standardize the enterprise's internal assessment data to generate classification data on changes in enterprise asset size, changes in enterprise business scope, changes in enterprise management, administrative penalty label classification data, and public opinion label classification data; the process of generating the classification data on changes in enterprise asset size includes:

[0030] Quantitatively extract the changes in the enterprise's asset size from the enterprise's internal valuation data to obtain customer throughput, current GDP and enterprise asset size data;

[0031] Converting the customer throughput, GDP at current prices, and enterprise asset size data into low-dimensional customer throughput data, GDP data, and enterprise asset data through log function conversion;

[0032] Performing z-score normalization on the customer throughput data, the GDP data, and the enterprise asset data to form enterprise asset scale change classification data;

[0033] Performing zero-one-hot encoding processing on the enterprise asset scale change classification data, the enterprise business scope change classification data, the enterprise management change classification data, the enterprise administrative penalty label classification data, and the enterprise public opinion label classification data to form time comparison analysis data;

[0034] Parameter extraction is performed on the time comparison analysis data to form internal reference data.

[0035] Optionally, obtaining primary reference information through the multivariate regression model, obtaining auxiliary reference information through the temporal difference autoregressive moving average model, obtaining a predicted customer throughput trend based on the primary reference information and the auxiliary reference information, and obtaining risk warning information of the current enterprise based on the customer throughput trend and preset trend risk comparison information include:

[0036] Obtain the weight of each factor in the multivariate regression model, and assign a stage value to each factor based on the weight of each factor to obtain the significance of each factor; obtain the macroeconomic trend and enterprise industry price trend in the preset quarter based on the time-dependent autoregressive moving average model;

[0037] The weights and significance of each factor are used as primary reference information, and the macroeconomic trend and enterprise industry price trend in the preset quarter are used as auxiliary reference information. The primary reference information and the auxiliary reference information are fitted and solved using a preset optimal model and a preset fitting algorithm to predict customer throughput trends;

[0038] The increase or decrease range of the throughput is obtained according to the throughput trend, and the increase or decrease range is matched with preset warning information to obtain risk warning information.

[0039] In order to solve the above problems, the present invention further provides an enterprise strategic intelligent early warning device, the device comprising:

[0040] A data acquisition unit, configured to retrieve external business data based on a preset database or data retrieval platform, focusing on fields related to pre-established external business environment themes, and retrieve internal enterprise assessment data based on fields related to pre-established internal assessment themes;

[0041] A feature extraction unit is used to perform natural language recognition processing on the external business data to extract external feature data; and perform standardization and coding processing on the internal evaluation data of the enterprise to obtain internal reference data;

[0042] An average regression unit is used to perform node averaging processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal difference autoregressive sliding average model; and to construct a regression model using the internal reference data as a factor to obtain a multivariate regression model;

[0043] A risk warning unit is used to obtain primary reference information through the multivariate regression model, obtain auxiliary reference information through the temporal difference autoregressive moving average model, obtain predicted customer throughput trends based on the primary reference information and the auxiliary reference information, and obtain risk warning information for the current enterprise based on the customer throughput trends and preset trend risk control information.

[0044] In order to solve the above problem, the present invention further provides an electronic device, comprising:

[0045] a memory storing at least one instruction; and

[0046] The processor executes the instructions stored in the memory to implement the steps in the above-mentioned enterprise strategic intelligent early warning method.

[0047] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned enterprise strategic intelligent early warning method.

[0048] The embodiment of the present invention first establishes an enterprise external business environment theme and an enterprise internal evaluation theme about a preset business, and based on a preset database or data retrieval platform, retrieves external business data around the external business environment theme and retrieves enterprise internal evaluation data around the internal evaluation theme; performs NLP recognition processing on the external business data to extract external feature data; performs standardization processing and coding processing on the enterprise internal evaluation data to obtain internal reference data; performs quarterly average processing on the external feature data by time series technology to obtain a temporal difference autoregressive sliding average model; constructs a regression model using the internal reference data as a factor to obtain a multiple regression model; obtains primary reference information through the multiple regression model, obtains auxiliary reference information through the temporal difference autoregressive sliding average model, and obtains a predicted customer throughput trend based on the primary reference information and the auxiliary reference information. And according to the customer throughput trend and the preset trend risk control information, the current enterprise risk warning information is obtained. In this way, the large customer risk warning model analyzes the business transaction data between large customers and port group companies, analyzes and monitors the fluctuation pattern of large customers' historical business data, and combines macroeconomic trends, large customers' industry policies, industry supply and demand trends, industry public opinion, large customers' enterprise scale and core personnel changes, corporate public opinion and other factors to analyze and evaluate the influencing factors of business data in different periods, build a risk factor feature library, and combine real-time change information and algorithm models to realize the warning of whether the customer may be at risk in the next cycle and the probability of occurrence, so as to realize intelligent warning of customer risks and reduce possible losses discovered afterwards. That is, it realizes intelligent warning of customer risks and reduces possible losses discovered afterwards; evaluates the accuracy and practicality of the model to improve the throughput prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flow chart of an enterprise strategic intelligent early warning method provided by one embodiment of the present invention;

[0050] Figure 2 A schematic diagram of a module of an enterprise strategic intelligent early warning device provided by an embodiment of the present invention;

[0051] Figure 3 A schematic diagram of the internal structure of an electronic device for an enterprise strategic intelligent early warning method provided by an embodiment of the present invention;

[0052] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0053] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] Due to historical operational experience and technological limitations, most groups currently rely primarily on customer visits to obtain key account information. However, irregular visit cycles and incomplete information collection severely impact key account risk management. In most cases, key account risk information is only received after the fact, resulting in a lack of awareness of customer situations, delayed responses, and a failure to provide early or timely warnings. Once risks occur, they can easily lead to significant losses for the company. Preemptive identification of customer business risk behaviors and timely strategy development are crucial to the healthy development of a company.

[0055] To solve the above problems, an embodiment of the present invention provides an enterprise strategic intelligent early warning method.

[0056] In this embodiment, the executing entity is the enterprise strategic intelligent early warning system of the entire server cluster. The enterprise strategic intelligent early warning system is integrated in the server cluster, that is, different modules of the enterprise strategic intelligent early warning system under the server cluster perform different operation steps respectively. Among them, the architecture of the server cluster includes multiple servers, and multiple cluster instances run under each server. Multiple scheduled tasks are stored under each cluster instance. In this way, the orderly execution of the scheduled task is achieved through the following steps.

[0057] It should be noted that embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results.

[0058] like Figure 1 As shown, in this embodiment, the enterprise strategic intelligent early warning method includes:

[0059] S1: Based on a preset database or data retrieval platform, retrieve external business data based on the fields of the pre-established external business environment theme, and retrieve the enterprise's internal assessment data based on the fields of the pre-established internal assessment theme;

[0060] S2: Performing natural language recognition processing on the external business data to extract external feature data; performing standardization and coding processing on the internal enterprise evaluation data to obtain internal reference data;

[0061] S3: performing node averaging processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal differential autoregressive sliding average model; constructing a regression model using the internal reference data as factors to obtain a multivariate regression model;

[0062] S4: Obtain primary reference information through the multivariate regression model, obtain auxiliary reference information through the temporal differential autoregressive sliding average model, obtain a predicted customer throughput trend based on the primary reference information and the auxiliary reference information, and obtain risk warning information of the current enterprise based on the customer throughput trend and preset trend risk control information.

[0063] exist Figure 1 In the illustrated embodiment, step S1 is a step of retrieving external business data based on a pre-established external business environment subject field and retrieving internal enterprise assessment data based on a pre-established internal assessment subject field based on a preset database or data retrieval platform. This process includes:

[0064] S11: Dividing the internal and external data of the preset business into an external data set and an internal data set; wherein the external data set includes at least macroeconomic data, industry price and industry development data, and media data; and the internal data set includes customer business volume data, basic enterprise information data, and enterprise operation data within the enterprise group;

[0065] S12: Establishing an external business environment theme and an internal assessment theme for the preset business, and marking the external business environment theme on the external data set and the internal assessment theme on the internal data set;

[0066] S13: crawling external business data around the fields of the external business environment theme, and retrieving the enterprise's internal evaluation data around the fields of the internal evaluation theme; the external business data includes macroeconomic environment analysis, policy impact analysis, and industry technology development public opinion impact analysis. Due to the large proportion of foreign trade in port business, the impact of the external environment includes both domestic and foreign macroeconomic and industry environments; the enterprise's internal evaluation data includes the basic status of the enterprise and its operation and management situation;

[0067] The process of obtaining the enterprise's internal assessment data includes:

[0068] S1311: Obtain enterprise asset size changes, business scope changes, management changes, administrative penalty labels, and public opinion labels through the preset enterprise assessment plug-in;

[0069] S1312: Assign values ​​to the changes in the enterprise's asset scale, business scope, management level, administrative penalty label, and public opinion label to obtain an internal enterprise evaluation data table;

[0070] S1313: performing traversal calculation on the enterprise internal evaluation data table using a preset evaluation algorithm to obtain enterprise internal evaluation data;

[0071] The process of obtaining the macroeconomic environment analysis includes:

[0072] S1321: Obtain the customer's throughput in the enterprise through a preset information crawling plug-in, and simultaneously obtain the current GDP price, average exchange rate, and import and export policy labels through a preset media information plug-in. In this embodiment, the enterprise is a port group, and the port's throughput is obtained.

[0073] S1322: Creating a table to be filled in according to a preset arrangement rule, and mapping the throughput, the current price of GDP, the average exchange rate, and the import and export policy label in the table to be filled in to form a macroeconomic environment analysis table;

[0074] S1323: Inputting the macroeconomic environment analysis table into a preset intelligent information extraction model, so that the intelligent information extraction model automatically outputs a macroeconomic environment analysis according to the macroeconomic environment analysis table.

[0075] In this embodiment, all data on customer business volume and possible influencing factors are obtained, including customer business volume data within the port group, external macroeconomic data, industry prices and industry development data, enterprise basic information data, enterprise operating data, related news, etc.

[0076] Taking the port group as an example, changes in the macroeconomic environment and business operations often have a delayed and non-instantaneous impact on the performance of large-scale import and export businesses. Therefore, the dependent variable is designed to be the customer throughput of the port group. Macroeconomic impacts are represented using current GDP prices, average exchange rates, and import and export policy labels. The impact of the industry environment includes industry price indices, industry (domestic / exporting country) policy labels, and industry (domestic / exporting country) news labels.

[0077] exist Figure 1 In the embodiment shown, step S2 is a process of performing natural language recognition processing on the external business data to extract external feature data; and performing standardization and encoding processing on the internal enterprise evaluation data to obtain internal reference data. This process includes:

[0078] The process of performing NLP recognition processing on the external business data to extract external feature data includes:

[0079] S211: performing algorithm fitting based on pre-acquired sample data about a preset business to obtain an NLP recognition model for starting an NLP semantic recognition service;

[0080] S212: Training the NLP recognition model with the sample data related to the preset business to obtain an NLP semantic recognition service process;

[0081] S213: Processing the external business data to form standard data, inputting the standard data into the NLP semantic recognition service process, and causing the NLP semantic recognition service process to perform semantic recognition on the external business data to obtain enterprise-related keywords and data corresponding to the keywords; wherein the process of processing the external business data to form standard data is to convert the external business data into a format in which one item corresponds to one data;

[0082] S214: performing synonym replacement on the keywords to obtain near-meaning keywords, and packaging the keywords, data corresponding to the keywords, and the near-meaning keywords to form external feature data.

[0083] The process of standardizing and coding the enterprise's internal assessment data to obtain internal reference data includes:

[0084] S221: Standardize the enterprise internal assessment data to generate enterprise asset size change classification data, enterprise business scope change classification data, enterprise management change classification data, enterprise administrative penalty label classification data, and enterprise public opinion label classification data; wherein, the process of generating enterprise asset size change classification data includes:

[0085] Quantitatively extract the changes in the enterprise's asset size from the enterprise's internal valuation data to obtain customer throughput, current GDP and enterprise asset size data;

[0086] Converting the customer throughput, GDP at current prices, and enterprise asset size data into low-dimensional customer throughput data, GDP data, and enterprise asset data through log function conversion;

[0087] Performing z-score normalization on the customer throughput data, the GDP data, and the enterprise asset data to form enterprise asset scale change classification data;

[0088] S222: performing zero-one-hot encoding processing on the enterprise asset scale change classification data, the enterprise business scope change classification data, the enterprise management change classification data, the enterprise administrative penalty label classification data, and the enterprise public opinion label classification data to generate time comparison analysis data;

[0089] S223: Extract parameters from the time comparison analysis data to form internal reference data.

[0090] Specifically, step S2 is feature engineering, including standardization of numerical data, NLP processing of text data, one-hot encoding of categorical data, feature selection and dimensionality reduction, etc. For example, numerical data such as customer throughput, current price of GDP, and enterprise asset size are converted using the log function to reduce the impact of dimension, and the converted data, average exchange rate, industry price index, etc. are subjected to z-score standardization for factor analysis; text data such as macro policies, industry policies, industry news, and corporate public opinion are subjected to word segmentation, synonym replacement, paragraph keyword extraction, document summary extraction, etc., and 5-8 frequently appearing keywords in each text are selected as feature data for each customer at different times; categorical data such as whether the business scope of the enterprise has changed, whether the management has changed, administrative penalty classification and whether it has been punished are 0-1 one-hot encoded; correlation analysis is performed on the processed data and customer throughput data to select features with large correlation and reasonableness; finally, panel data for different customers in each quarter are formed.

[0091] exist Figure 1 In the embodiment shown, step S3 is a process of performing quarterly averaging processing on the external characteristic data by using a time series technique to obtain a temporal difference autoregressive moving average model; and constructing a regression model using the internal reference data as factors to obtain a multiple regression model.

[0092] In a specific embodiment, taking the port group as an enterprise as an example, time series technology is used to construct a SARIMA (temporal autoregressive moving average model) for external characteristic data such as macroeconomic GDP and industry price index to predict short-term (2 quarters) macro trends and industry price trends as a reference for decision-making auxiliary information; b. A multivariate regression model is constructed step by step for the characteristic data of internal reference data such as customer throughput data and feature-processed GDP current price, average exchange rate, industry price index, macro policy keywords, industry news keywords, corporate public opinion keywords, enterprise scale changes, whether the enterprise management changes, and whether the enterprise has various administrative penalties for the first, second, and third periods (step-by-step regression model steps: different factors at different lag periods are gradually added to the regression model. If the model fit is significantly improved and the factor coefficient is estimated to be 90% significant, the factor is retained; otherwise, it is not retained; the model fit is reconstructed and compared with different factor addition orders to select the optimal model and factors).

[0093] exist Figure 1 In the illustrated embodiment, step S4 is to obtain primary reference information through the multivariate regression model, obtain auxiliary reference information through the temporal difference autoregressive moving average model, obtain a predicted customer throughput trend based on the primary and auxiliary reference information, and obtain risk warning information for the current enterprise based on the customer throughput trend and preset trend risk comparison information. This includes:

[0094] S41: Obtaining the weight of each factor in the multivariate regression model, and assigning a stage value to each factor according to the weight of each factor to obtain the significance of each factor; obtaining the macroeconomic trend and the enterprise industry price trend in a preset quarter according to the time-dependent autoregressive moving average model;

[0095] S42: Using the weights and significance of each factor as primary reference information, and the macroeconomic trend and enterprise industry price trend in the preset quarter as auxiliary reference information, and fitting and solving the primary reference information and the auxiliary reference information using a preset optimal model and a preset fitting algorithm to predict customer throughput trends;

[0096] S43: Obtaining a throughput increase or decrease range according to the throughput trend, and comparing the throughput increase or decrease range with preset warning information to obtain risk warning information;

[0097] Among them, low, medium and high risk warning information will be set for predicted throughput reductions exceeding 10%, 30% and 50% respectively;

[0098] Specifically, in this embodiment, the weight and significance of each influencing factor are first analyzed, and then the optimal model is used to predict the customer throughput trend based on the latest data of these factors and the data predicted in step a. Low, medium and high risk warning information are set for predicted throughput reductions exceeding 10%, 30% and 50%, respectively; in addition, data such as corporate changes, news, public opinion, etc. with keyword combinations with high model influence weights can be popped up as decision-making auxiliary information reference to obtain final risk information.

[0099] More specifically, this embodiment aims to study whether there is a risk of a large-scale reduction in the business volume of port enterprises. The analysis path is whether the client has a large change in business volume. Based on the experience of corporate management theory and business practice, there are two types of factors that affect the client's business performance: one is the external business environment in which the enterprise is located, including the macroeconomic environment, policy influences, and the technology, development, public opinion, etc. of the industry in which it is located. Due to the large proportion of foreign trade in port business, the impact of the external environment includes both domestic and foreign macroeconomic and industry environments; the other is the inherent quality of the enterprise itself, that is, the basic conditions and business management of the enterprise; in addition, changes in the macro-environment and corporate operations are often reflected in the business performance of large-scale import and export businesses with a delayed and non-immediate response. Based on this, the dependent variable characteristics are designed to be the customer throughput of this port group; the impact of the macroeconomy is represented by GDP at current prices, average exchange rates, import and export policy labels, etc.; the impact of the industry environment includes industry price index, industry (domestic / exporting country) policy labels, and industry (domestic / exporting country) news labels; the impact of corporate operations includes changes in corporate asset scale, changes in corporate business scope, changes in corporate management, corporate administrative penalty labels, corporate public opinion labels, etc. Based on this, risk information is obtained according to the proportion of each information to improve the rigor and comprehensiveness of risk assessment.

[0100] In this embodiment, the server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0101] As described above, the enterprise strategy intelligent early warning method provided by the embodiment of the present invention establishes an enterprise external business environment theme and an enterprise internal evaluation theme about a preset business, retrieves external business data around the external business environment theme, and retrieves the enterprise internal evaluation data around the internal evaluation theme; performs NLP recognition processing on the external business data to extract external feature data; performs standardization processing and coding processing on the enterprise internal evaluation data to obtain internal reference data, and performs quarterly average processing on the external feature data by quarter through time series technology to obtain a temporal difference autoregressive sliding average model; uses the internal reference data as a factor to construct a regression model to obtain a multiple regression model, then obtains primary reference information through the multiple regression model, obtains auxiliary reference information through the temporal difference autoregressive sliding average model, and obtains a predicted customer throughput trend based on the primary reference information and the auxiliary reference information, and obtains the risk information of the current port based on the preset trend risk control information. This can achieve intelligent early warning of customer risks, reduce possible losses discovered afterwards, and can evaluate the accuracy and practicality of the model to improve the throughput prediction accuracy.

[0102] like Figure 2 As shown, the present invention provides an enterprise strategic intelligent early warning device 100, which can be installed in an electronic device. Depending on the functions to be implemented, the enterprise strategic intelligent early warning device 100 may include a data acquisition unit 101, a feature extraction unit 102, a mean regression unit 103, and a risk early warning unit 104. The module described in the present invention, also referred to as a unit, refers to a series of computer program segments that can be executed by an electronic device processor and can perform a fixed function, which is stored in the memory of the electronic device.

[0103] In this embodiment, the functions of each module / unit are as follows:

[0104] The data acquisition unit 101 is used to retrieve external business data based on a preset external business environment theme field and retrieve enterprise internal evaluation data based on a preset internal evaluation theme field based on a preset database or data retrieval platform;

[0105] The feature extraction unit 102 is configured to perform natural language recognition processing on the external business data to extract external feature data; and perform standardization and coding processing on the internal enterprise evaluation data to obtain internal reference data;

[0106] The average regression unit 103 is configured to perform node average processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal difference autoregressive moving average model; and construct a regression model using the internal reference data as factors to obtain a multivariate regression model;

[0107] The risk warning unit 104 is used to obtain primary reference information through the multivariate regression model, obtain auxiliary reference information through the temporal difference autoregressive moving average model, obtain predicted customer throughput trends based on the primary reference information and the auxiliary reference information, and obtain risk warning information of the current enterprise based on the customer throughput trends and preset trend risk control information.

[0108] As described above, the enterprise strategy intelligent early warning device 100 provided by the present invention first sets up an enterprise external business environment theme and an enterprise internal evaluation theme about a preset business based on the data acquisition unit 101, retrieves external business data around the external business environment theme, and retrieves enterprise internal evaluation data around the internal evaluation theme; then performs NLP recognition processing on the external business data through the feature extraction unit 102 to extract external feature data; performs standardization processing and coding processing on the enterprise internal evaluation data to obtain internal reference data, and then performs mean regression processing on the external feature data by quarter through time series technology. Quarterly average processing is performed to obtain a temporal differential autoregressive moving average model; the internal reference data is used as a factor to construct a regression model to obtain a multiple regression model, and finally, based on the risk warning unit 104, the main reference information is obtained through the multiple regression model, and the auxiliary reference information is obtained through the temporal differential autoregressive moving average model, and the customer throughput trend is predicted based on the main reference information and the auxiliary reference information, and the risk information of the current port is obtained based on the preset trend risk control information. This can achieve intelligent early warning of customer risks, reduce possible losses discovered afterwards, and can evaluate the accuracy and practicality of the model, thereby improving the throughput prediction accuracy.

[0109] like Figure 3 As shown, the present invention provides an electronic device 1 for an enterprise strategic intelligent early warning method.

[0110] The electronic device 1 may include a processor 10 , a memory 11 and a bus, and may also include a computer program stored in the memory 11 and executable on the processor 10 , such as an enterprise strategic intelligent early warning program 12 .

[0111] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (for example, SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Furthermore, the memory 11 can also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed on the electronic device 1, such as the code of the enterprise strategic intelligent early warning program, but can also be used to temporarily store data that has been output or is to be output.

[0112] In some embodiments, the processor 10 may be composed of an integrated circuit, such as a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips. The processor 10 is the control core (Control Unit) of the electronic device, connecting the various components of the entire electronic device using various interfaces and lines, and executing programs or modules stored in the memory 11 (such as an enterprise strategic intelligent early warning program, etc.), as well as calling data stored in the memory 11, to perform various functions of the electronic device 1 and process data.

[0113] The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable connection and communication between the memory 11 and at least one processor 10, etc.

[0114] Figure 3 Only the electronic device with components is shown, and it can be understood by those skilled in the art that Figure 3The structure shown does not constitute a limitation on the electronic device 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0115] For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for powering the various components. Preferably, the power source may be logically connected to the at least one processor 10 via a power management device, thereby implementing functions such as charging management, discharging management, and power consumption management through the power management device. The power source may further include any components such as one or more DC or AC power sources, a recharging device, a power failure detection circuit, a power converter or inverter, a power status indicator, etc. The electronic device 1 may further include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0116] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0117] Optionally, the electronic device 1 may further include a user interface, which may be a display or an input unit (such as a keyboard). Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0118] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0119] The enterprise strategic intelligent early warning program 12 stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve the following:

[0120] Based on a preset database or data retrieval platform, external business data is retrieved based on fields related to pre-established external business environment themes, and internal enterprise assessment data is retrieved based on fields related to pre-established internal assessment themes;

[0121] Performing natural language recognition processing on the external business data to extract external feature data; performing standardization and coding processing on the internal enterprise evaluation data to obtain internal reference data;

[0122] Performing node averaging processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal differential autoregressive sliding average model; constructing a regression model using the internal reference data as factors to obtain a multivariate regression model;

[0123] The primary reference information is obtained through the multivariate regression model, the auxiliary reference information is obtained through the temporal differential autoregressive sliding average model, the customer throughput trend is predicted based on the primary reference information and the auxiliary reference information, and the risk warning information of the current enterprise is obtained based on the customer throughput trend and preset trend risk control information.

[0124] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figure 1 The description of the relevant steps in the corresponding embodiment is not repeated here. It should be emphasized that in order to further ensure the privacy and security of the above-mentioned enterprise strategic intelligent warning, the data of the above-mentioned enterprise strategic intelligent warning is stored in the node of the blockchain where this server cluster is located.

[0125] The server can be a stand-alone server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), as well as big data and artificial intelligence platforms.

[0126] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0127] An embodiment of the present invention further provides a computer-readable storage medium, which may be non-volatile or volatile, and stores a computer program. When the computer program is executed by a processor, the following is achieved:

[0128] Based on a preset database or data retrieval platform, external business data is retrieved based on fields related to pre-established external business environment themes, and internal enterprise assessment data is retrieved based on fields related to pre-established internal assessment themes;

[0129] Performing natural language recognition processing on the external business data to extract external feature data; performing standardization and coding processing on the internal enterprise evaluation data to obtain internal reference data;

[0130] Performing node averaging processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal differential autoregressive sliding average model; constructing a regression model using the internal reference data as factors to obtain a multivariate regression model;

[0131] The primary reference information is obtained through the multivariate regression model, the auxiliary reference information is obtained through the temporal differential autoregressive sliding average model, the customer throughput trend is predicted based on the primary reference information and the auxiliary reference information, and the risk warning information of the current enterprise is obtained based on the customer throughput trend and preset trend risk control information.

[0132] Specifically, the specific implementation method when the computer program is executed by the processor can refer to the description of the relevant steps in the enterprise strategic intelligent early warning method in the embodiment, and will not be repeated here.

[0133] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the module division is merely a logical function division, and other division methods may be used in actual implementation.

[0134] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0135] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0136] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0137] Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the claims are intended to be embraced therein. Any reference to a figure in a claim should not be construed as limiting the claim to which it relates.

[0138] Blockchain, as used in this article, refers to a novel application model for computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Blockchain is essentially a decentralized database, a series of data blocks generated using cryptographic methods. Each block contains information about a batch of online transactions, used to verify the validity of this information (to prevent counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, the platform product service layer, and the application service layer.

[0139] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. Second-order terms are used to indicate names and do not imply any particular order.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An enterprise strategic intelligent early warning method, characterized in that: Based on artificial intelligence technology, including: Based on a preset database or data retrieval platform, external business data is retrieved around the fields of a preset external business environment theme, and internal enterprise evaluation data is retrieved around the fields of a preset internal evaluation theme; wherein, the method includes: dividing the internal and external data of the preset business into an external data set and an internal data set; wherein, the external data set includes at least macroeconomic data, industry price and industry development data, and media data; the internal data set includes customer business volume data, enterprise basic information data, and enterprise business data within the enterprise group; establishing an enterprise external business environment theme and an enterprise internal evaluation theme for the preset business, and marking the enterprise external business environment theme on the external data set, and marking the enterprise internal evaluation theme on the internal data set; crawling external business data around the fields of the external business environment theme, and retrieving enterprise internal evaluation data around the fields of the internal evaluation theme; Performing natural language recognition processing on the external business data based on the NLP recognition model for the NLP semantic recognition service to extract external feature data; performing standardization and encoding processing on the internal enterprise evaluation data to obtain internal reference data; Performing node averaging processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal differential autoregressive sliding average model; constructing a regression model using the internal reference data as factors to obtain a multivariate regression model; Obtaining primary reference information through the multivariate regression model, obtaining auxiliary reference information through the temporal difference autoregressive moving average model, obtaining a predicted customer throughput trend based on the primary and auxiliary reference information, and obtaining risk warning information for the current enterprise based on the customer throughput trend and preset trend risk comparison information; this includes: obtaining weights of various factors in the multivariate regression model, assigning stage values ​​to each factor based on the weights of the various factors to obtain the significance of each factor; and obtaining macroeconomic trends and enterprise industry price trends in a preset quarter based on the temporal difference autoregressive moving average model; The weights and significance of each factor are used as primary reference information, and the macroeconomic trends and enterprise industry price trends in the preset quarter are used as auxiliary reference information. The primary reference information and the auxiliary reference information are fitted and solved using a preset optimal model and a preset fitting algorithm to predict customer throughput trends.

2. The enterprise strategic intelligent early warning method according to claim 1, characterized in that: The external business data at least includes macroeconomic environment analysis, policy impact analysis and industry technology development public opinion impact analysis.

3. The enterprise strategic intelligent early warning method according to claim 2, characterized in that: The retrieval of the enterprise's internal assessment data includes: Obtain enterprise asset size changes, business scope changes, management changes, administrative penalty labels, and public opinion labels through pre-set enterprise assessment plug-ins; Assign values ​​to the changes in the enterprise's asset scale, business scope, management level, administrative penalty label, and public opinion label to obtain an internal assessment data table for the enterprise; The enterprise internal evaluation data table is traversed and calculated using a preset evaluation algorithm to obtain the enterprise internal evaluation data.

4. The enterprise strategic intelligent early warning method according to claim 2, characterized in that: Obtain analysis of the macroeconomic environment, including: The customer's throughput in the enterprise is obtained through the preset information crawling plug-in, and the current GDP price, average exchange rate, and import and export policy labels are obtained through the preset media information plug-in; Creating a table to be filled in according to a preset arrangement rule, and mapping the throughput, the current price of GDP, the average exchange rate, and the import and export policy label in the table to be filled in to form a macroeconomic environment analysis table; The macroeconomic environment analysis table is input into a preset intelligent information extraction model, so that the intelligent information extraction model automatically outputs a macroeconomic environment analysis according to the macroeconomic environment analysis table.

5. The enterprise strategic intelligent early warning method according to claim 1, characterized in that: The performing natural language recognition processing on the external business data to extract external feature data includes: Performing algorithm fitting based on pre-acquired sample data about a preset business to obtain an NLP recognition model for starting the NLP semantic recognition service; Based on the NLP recognition model, training is performed using the sample data related to the preset business to obtain an NLP semantic recognition service process; The external business data is processed to form standard data, and the standard data is input into the NLP semantic recognition service process, so that the NLP semantic recognition service process performs semantic recognition on the external business data to obtain enterprise-related keywords and data corresponding to the keywords; wherein the process of processing the external business data to form standard data is to change the external business data into a format in which one item corresponds to one data; Synonym replacement is performed on the keywords to obtain near-meaning keywords, and the keywords, data corresponding to the keywords, and the near-meaning keywords are packaged to form external feature data.

6. The enterprise strategic intelligent early warning method according to claim 5, characterized in that: The standardization and coding of the enterprise internal evaluation data to obtain internal reference data includes: Standardize the enterprise's internal assessment data to generate classification data on changes in enterprise asset size, changes in enterprise business scope, changes in enterprise management, administrative penalty label classification data, and public opinion label classification data; the process of generating the classification data on changes in enterprise asset size includes: Quantitatively extract the changes in the enterprise's asset size from the enterprise's internal valuation data to obtain customer throughput, current GDP and enterprise asset size data; Converting the customer throughput, GDP at current prices, and enterprise asset size data into low-dimensional customer throughput data, GDP data, and enterprise asset data through log function conversion; Performing z-score normalization on the customer throughput data, the GDP data, and the enterprise asset data to form enterprise asset scale change classification data; Performing zero-one-hot encoding processing on the enterprise asset scale change classification data, the enterprise business scope change classification data, the enterprise management change classification data, the enterprise administrative penalty label classification data, and the enterprise public opinion label classification data to form time comparison analysis data; Parameter extraction is performed on the time comparison analysis data to form internal reference data.

7. The enterprise strategic intelligent early warning method according to claim 6, characterized in that: Obtaining risk warning information of the current enterprise based on the customer throughput trend and preset trend risk comparison information, including: The increase or decrease range of the throughput is obtained according to the throughput trend, and the increase or decrease range is matched with preset warning information to obtain risk warning information.

8. An enterprise strategic intelligent early warning device, characterized in that: The device is based on artificial intelligence technology and includes: A data acquisition unit is used to retrieve external business data around the fields of a pre-established external business environment theme, and retrieve enterprise internal evaluation data around the fields of a pre-established internal evaluation theme based on a preset database or data retrieval platform; wherein, the data includes: dividing the internal and external data of the preset business into an external data set and an internal data set; wherein, the external data set includes at least macroeconomic data, industry price and industry development data, and media data; the internal data set includes customer business volume data, enterprise basic information data, and enterprise business data within the enterprise group; establishing an enterprise external business environment theme and an enterprise internal evaluation theme for the preset business, and marking the enterprise external business environment theme on the external data set and the enterprise internal evaluation theme on the internal data set; crawling external business data around the fields of the external business environment theme, and retrieving enterprise internal evaluation data around the fields of the internal evaluation theme; a feature extraction unit configured to perform natural language recognition processing on the external business data based on an NLP recognition model for an NLP semantic recognition service to extract external feature data; and perform standardization and encoding processing on the enterprise internal evaluation data to obtain internal reference data; An average regression unit is used to perform node averaging processing on the external feature data according to preset time nodes through a preset time series to obtain a temporal difference autoregressive sliding average model; and to construct a regression model using the internal reference data as a factor to obtain a multivariate regression model; a risk warning unit configured to obtain primary reference information through the multivariate regression model, obtain auxiliary reference information through the temporal difference autoregressive moving average model, obtain a predicted customer throughput trend based on the primary and auxiliary reference information, and obtain risk warning information for the current enterprise based on the customer throughput trend and preset trend risk comparison information; the unit includes: obtaining weights of various factors in the multivariate regression model, assigning stage values ​​to each factor based on the weights of the various factors to obtain the significance of the various factors; and obtaining macroeconomic trends and enterprise industry price trends in a preset quarter based on the temporal difference autoregressive moving average model; The weights and significance of each factor are used as primary reference information, and the macroeconomic trends and enterprise industry price trends in the preset quarter are used as auxiliary reference information. The primary reference information and the auxiliary reference information are fitted and solved using a preset optimal model and a preset fitting algorithm to predict customer throughput trends.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the steps in the enterprise strategic intelligent early warning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the enterprise strategic intelligent early warning method as described in any one of claims 1 to 7 is implemented.

Images