CN113780604A - Composite enterprise credit early warning system and method - Google Patents
Composite enterprise credit early warning system and method Download PDFInfo
- Publication number
- CN113780604A CN113780604A CN202010442991.4A CN202010442991A CN113780604A CN 113780604 A CN113780604 A CN 113780604A CN 202010442991 A CN202010442991 A CN 202010442991A CN 113780604 A CN113780604 A CN 113780604A
- Authority
- CN
- China
- Prior art keywords
- enterprise
- module
- financial
- public opinion
- credit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/358—Browsing; Visualisation therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/12—Accounting
- G06Q40/125—Finance or payroll
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Finance (AREA)
- General Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Technology Law (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention provides a composite enterprise credit early warning system and a method; the method solves the problems in the conventional enterprise credit rating and early warning, and improves the timeliness and comprehensiveness of credit analysis and early warning; the method takes a negative public opinion monitoring mechanism as a core, and simultaneously combines a comprehensive credit risk early warning platform applied by market price, financial data and the like, so that default risk points of debt main bodies can be monitored in a coordinated and complementary mode in all directions, and risk early warning is carried out in a full time period; the system comprises a negative public opinion module, a default probability module, a financial rating module and a comprehensive analysis module; the system can be used for timely and accurately early warning the credit of the target enterprise.
Description
Technical Field
The invention relates to the field of financial science and technology, in particular to a composite enterprise credit early warning system and a composite enterprise credit early warning method.
Background
At present, the credit analysis and early warning of enterprises or institutions in the Chinese financial industry are mainly realized by monitoring the credit rating and the change thereof. Credit rating is usually provided by external professional institutions (such as rating companies) and internal research departments of financial institutions, and the core idea is to comprehensively analyze financial data of the companies, competitive advantages of enterprises, influence of industrial and macroscopic policies, liability payment capacity and potential default loss of the enterprises, give relative ranking, and assign rating marks, such as AAA, AA, a, BBB, BB, B, etc., to distinguish risk levels according to historical experience and credit default cases. External and internal rating agencies may monitor market and business changes to adjust the credit rating of the business. And the investor adjusts the investment strategy according to the change of the rating.
The main factor influencing the credit rating of the enterprise is the change of the operation condition, industry and market of the enterprise. The financial data on which external ratings and internal ratings can depend is the company's regular public financial disclosure, such as quarterly or annually. In general, disclosure of financial data is delayed because the consolidation of corporate financial data requires time. As reported by the listed companies in years, the expiration date is typically the end of the next year, April. In addition, due to resource limitations, rating agencies (including internal rating departments of external rating companies and investment institutions) cannot track financial disclosures of enterprises in a timely manner to adjust ratings. Indeed, even in the international mature capital market, institutional investors are generally aware that adjustments to ratings often lag the occurrence of serious credit events such as violations. In the financial market of China, the reason for the conventional credit rating lagging the credit event and the market reaction is that besides the defect of the rating method, the reason can also be attributed to that enterprises do not comply with the corresponding specifications in the information disclosure and even have misleading and false actions. In addition, the lack of default cases, especially historical financial data of default enterprises, is often not disclosed, and the credit rating model cannot be subjected to relatively accurate parameter calibration, so that the performance of the model is seriously influenced. More importantly, in the enterprise operation process, important events caused by internal or external reasons, which occur between disclosure time points, are generally reported by financial media, and the influence of the events on the credit rating of the enterprise can be generally only interpreted and analyzed by rating personnel to determine whether to adjust the rating. Similarly, due to resource limitations, changes in enterprise credit caused by unscheduled information disclosure often do not reflect in time in the adjustment of credit ratings.
China's financial institutions, particularly small and medium-sized financial institutions including banks, security dealers, fund companies, insurance companies and the like, generally need to be able to timely monitor credit changes of the whole financial market and specific enterprises, including credit changes embodied in regular information disclosure and implied in daily financial media reports.
Disclosure of Invention
The method aims to solve the problems in conventional enterprise credit rating and early warning, improve the timeliness of credit analysis and early warning, and reflect the financial data change displayed in regular information disclosure in time and the influence of enterprise operation condition change implied in daily financial reports on the credit change; comprehensive, fully covering credit analysis of capital markets, particularly corporate bonds traded in open markets; accuracy, improve early warning ability. This patent uses negative public opinion monitoring mechanism as the core, combines the comprehensive credit risk early warning platform of applications such as market price, financial data simultaneously, can cooperate, the comprehensive monitoring debt main part default risk point of complementary ground, carries out the risk early warning for the full time quantum.
This patent includes the following:
a composite enterprise credit early warning system comprises a negative public opinion module, a default probability module, a financial rating module and a comprehensive analysis module;
the negative public opinion module comprises an enterprise public opinion information acquisition system, an enterprise negative public opinion processing system and an enterprise negative public opinion display system;
the default probability module comprises an enterprise security information acquisition system, an enterprise default probability prediction system and a default probability result display system;
the financial rating module comprises an enterprise comprehensive financial information acquisition system, an enterprise financial analysis system and a financial rating result display system;
and the comprehensive analysis module comprehensively analyzes the analysis results of the negative public opinion module, the default probability module and the financial rating module and then displays the whole result.
Preferably, the negative public opinion module in the composite enterprise credit early warning system, the enterprise public opinion information acquisition system comprises a unit for acquiring a target enterprise public opinion text through a crawler technology and a text interface;
the enterprise negative public opinion processing system comprises an enterprise public opinion text analysis unit and a processing unit;
the enterprise negative public opinion display system displays the enterprise negative public opinions obtained through comprehensive analysis by a theme accumulated line graph or a theme radar graph.
Preferably, the enterprise public opinion text analysis unit in the composite enterprise credit early warning system performs natural language analysis on the enterprise public opinion text, including emotion classification technology, to obtain basic language features, and the basic language features are used for evaluating adverse effects on enterprise credit to obtain a theme cumulative line graph or a theme radar graph of the enterprise negative public opinion.
Preferably, the topic classification of the negative public opinion of the enterprise in the composite enterprise credit early warning system includes: ability, willingness, condition, compliance, market, high management, and other negative public opinion information for businesses outside of the six categories mentioned above.
The negative public opinion monitoring mechanism of this patent uses artificial intelligence analytical technique (AI), has adopted present advanced semantic analysis, natural language processing and techniques such as emotion classification comprehensively, can be to the accurate analysis of all kinds of news through semi-supervised learning mode. The problem of most public opinion information monitoring system often attach importance to news quantity and ignore news quality on the market, blind propelling movement magnanimity news information leads to similar news repeatability higher, often leads to the early warning of inefficiency or mistake to report to the problem to and the user snatchs the news main points difficulty, receives that irrelevant news interference is great, is easily misled by irrelevant news. The negative public opinion monitoring mechanism of this patent utilizes the AI technique enough in time snatchs, rationally classifies and the analysis to magnanimity negative news, extracts the investor and thinks that the negative news related with the main part risk of default, improves the efficiency that the user reviewed greatly.
The information source of the negative public opinion monitoring mechanism of the patent comprises official websites, traditional media, mainstream financial media, internet social contact, platform websites and the like of a financial market. The specific information is classified by bulletin, news and WeChat, wherein the bulletin type includes illegal violation, information disclosure, transaction abnormity, rating change, important matters, advance payment, board change and the like, and the channel comes from the official website of the financial market, such as an exchange, a deep exchange, a foreign exchange transaction center, a medium debt registration, a clearing post and the like; the news covers reliable and the main news portal with high news quality authority, such as a news-mixing network, a Xinfeng financial channel, a phoenix financial channel and the like; the WeChat contains public numbers related to media bonds, such as dollar bond observation, medium integrity certificate evaluation, dealer China and the like.
According to the negative public opinion monitoring system, different crawler frequencies are adopted according to different types of websites, and crawling within 10s-180s of main stream hot spot media is guaranteed. All inquired information in the system is pushed in real time at a minute level, wherein the information can be simultaneously pushed in real time in the whole process of crawling, analyzing, storing and pushing the public opinion information, and 7-24 all-weather uninterrupted.
The processing process of the public opinion module is carried out concurrently, and the processing of news is mutually independent; the public opinion module is provided with an efficient analysis engine, each news analysis is completed within the second level, and the whole analysis process comprises similar text calculation, classification, main correlation calculation, summary acquisition, news scoring and the like; actively pushing public opinion results in real time; the public opinion processing server has multiple safety guarantees.
In the aspect of public opinion information mining analysis, the public opinion early warning system of the patent utilizes machine learning to label public opinion data based on semantic analysis technology. And one part of labeling is realized by learning historical data through a machine learning technology, nearly ten thousand samples are labeled, and high-quality data samples are provided for the model. And in part by rules implemented by experts configuring keywords (e.g., president lost, ran, listed as a deceased). The high automation is realized through artificial intelligence and an expert rule configuration mechanism, and the fastest information processing speed is achieved on the premise of ensuring the accuracy. The Hengtai intelligent public opinion analysis functions comprise news emotion analysis, theme classification, importance viewpoint, news relevancy, news score and the like.
And (3) news emotion analysis: it is determined whether the news is credit news (negative) or non-credit news (positive, neutral).
And (3) topic classification: the news is judged to belong to categories, wherein the categories comprise repayment willingness, repayment capacity, market development profiles, credit status, laws and regulations, high governance and other 7 subjects, and other subjects comprise subjects which are related to negative public sentiments of enterprises and are out of the six categories, such as sudden catastrophic events, industrial crisis, influence of sudden other bad events on the industries, long-lost contacts of the enterprises, roadways, distrusters listed as distrusters and the like.
News relevance: it is determined which subject the news is saying and the subject is determined about the event. The effect of simultaneous existence of multiple text bodies is more obvious.
Importance points: and similar to the abstract, the main body and the event are arranged.
And news scoring: and judging the score of the news corresponding to the main body, and judging the news with high risk, medium risk, low risk and no risk.
And (3) associating the map: and (4) carrying out atlas analysis on related subjects such as stockholders, mother and child companies, guarantors, investors and the like of the subjects.
Preferably, in the default probability module in the composite enterprise credit warning system, the enterprise security information acquisition system comprises an acquisition unit of target enterprise security information;
the enterprise default probability prediction system comprises an enterprise security information analysis unit and a processing unit; and the default probability result display system displays the default probability result through the EDF value.
Preferably, the enterprise security information analysis unit and the processing unit in the composite enterprise credit early warning system draw an enterprise asset value time curve by analyzing based on Merton option pricing as a theoretical basis and based on stock price quotation and financial newspaper information captured in the data acquisition unit, and a certain future time limit is given, so that the probability EDF value that the enterprise asset value is smaller than the default point is predicted.
The market information of the stocks and the bonds is often related to the situation of the issuing main body and the related bonds, and the dynamic trend of the market reflects different risk levels. The enterprise default probability prediction system of the patent incorporates securities market information, successfully realizes the prediction of default probability of a distribution subject in a default probability prediction model, and assists an investor to gain insight into bond default conditions by utilizing the credit profit difference analysis function of bonds.
>EDF
Based on Merton option pricing as a theoretical basis, stock price quotations, financial reports and the like as data bases, a certain future deadline (generally 1 year) is given, the probability that the asset value of an enterprise is smaller than a default point is predicted, and then the default possibility of the enterprise is known.
The system supports the analysis of the EDF value and the trend of an enterprise in nearly 8 years per day, and is expanded to the industry and regional analysis. The method can map the listed companies which stop the card for a long time and the non-listed companies which do not have stock price information to obtain results.
Difference of credit
By utilizing the bond market information and according to the change of the bond credit interest difference, the credit risk change of a main body corresponding to the bond can be reflected in time, and a complementary prediction effect is achieved when the EDF fails to reveal the default risk in advance.
Preferably, in the financial rating module of the composite enterprise credit warning system, the enterprise comprehensive financial information acquisition system comprises a text acquisition system for target enterprise comprehensive financial information through a crawler technology;
the enterprise comprehensive financial rating system comprises a text analysis unit and a processing unit;
the comprehensive financial rating result display system displays the rating result of the comprehensive financial of the enterprise through A +, A, A-and the operating income, net profit and net cash flow of the enterprise through a histogram.
Preferably, a text analysis unit and a processing unit in the composite enterprise credit early warning system score each financial index, index conversion and index weight of an enterprise by using a scoring card rating mechanism based on a Logistic model, then perform preliminary rating and model external adjustment to obtain final rating;
the method comprises the steps of establishing a general industrial and commercial enterprise, a city investment enterprise and a real estate characteristic rating model, establishing the significance and the correlation coefficient of related quantitative financial indexes and default of a target enterprise, and evaluating the credit rating of the target enterprise in the last 5 years.
The credit rating method can perform credit rating on all debt enterprises with financial information disclosure, and has the highest coverage on monitoring objects; the comprehensive financial rating model of this patent is based on the scoring card rating mechanism of Logistic model, brings into general industry and commerce enterprise, city and throws enterprise and real estate characteristic rating model, and the significance of the relevant quantitative financial index of debt subject to default and relevance between them is studied in depth, carries out the credit rating evaluation of nearly 5 years to the subject.
Preferably, the composite enterprise credit early warning system in the composite enterprise credit early warning system further comprises a target enterprise input system and a comprehensive collection early warning result display system;
the target enterprise input system inputs a target enterprise to be inquired through setting a dialog box, and enters the composite enterprise credit early warning system for operation;
the comprehensive collection early warning result display system comprehensively displays the results of the negative public opinion module, the default probability module and the comprehensive financial module through a table.
A composite enterprise credit early warning method comprises the following steps:
the method comprises the following steps: inputting a target enterprise to be searched through a target enterprise input system dialog box;
step two: triggering a negative public opinion module, a default probability module and a financial rating module of the composite enterprise credit early warning system through the name of a target enterprise, wherein the enterprise public opinion information acquisition system captures enterprise public opinion information, the enterprise security information acquisition system captures security information texts of the enterprise, and the enterprise comprehensive financial information acquisition system captures comprehensive financial information texts of the enterprise;
step three: the enterprise negative public opinion processing system, the enterprise default probability prediction system and the enterprise comprehensive financial rating system respectively analyze and process the captured enterprise negative public opinion information text, the enterprise security information text and the enterprise comprehensive financial information text to obtain respective analysis results;
step four: respectively displaying the results of each module in the third step through a result output unit, and performing comprehensive early warning display on the results through a comprehensive negative public opinion module, a default probability module and a financial rating module; the display platform terminal equipment comprises a mobile phone, a tablet personal computer and a desktop computer.
This patent uses three big credit risk management mechanisms of enterprise as the credit risk early warning platform of core, combines the advantage of three big credit risk management mechanisms, can cooperate, complementally all-round monitoring debt main part default risk point, carries out the risk early warning in the full time quantum.
The early warning modes of high timeliness and high coverage are different from the limitations brought by other single early warning models, the system collects three advanced credit risk management mechanisms, and the problem that fish and bear palms cannot be obtained at the same time is solved.
Negative public opinion monitoring: the method can capture external sudden negative news information in real time, and has the highest timeliness;
comprehensive financial rating: credit rating can be carried out on the debt enterprises with the financial information disclosure, and the coverage of the monitoring object is the highest;
and (3) default probability prediction: updating the default probability of the enterprise according to the security information of the enterprise every day, wherein the early warning frequency is high; the prediction of the enterprises on the market has higher accuracy and higher coverage.
Early warning grade under the comprehensive early warning scoring system:
the system finally shows the severity of the default risk in four grades of red, orange, yellow and green respectively. But behind these four levels there are a level of strict scoring criteria. And giving corresponding scores to the results obtained by each model according to a scientific scoring system, and finally obtaining the comprehensive early warning score by using a scene mathematical model. And ensuring that the slight change of any model result can be reflected on the comprehensive early warning score.
Advanced artificial intelligence techniques
The system uses artificial intelligence Analysis (AI) technology in the overall design and the construction of each large model, particularly in the negative public opinion monitoring process, the current advanced semantic analysis, natural language processing, emotion classification and other technologies are comprehensively adopted, and various news can be accurately analyzed through a semi-supervised learning mode. Most public opinion information monitoring systems in the market usually attach importance to news quantity and ignore news quality, and blindly push massive news information, so that the repeatability of similar news is higher, and inefficient or wrong early warning delivery is usually caused. The user is difficult to grasp news key points, has high interference by irrelevant news and is easily misled by the irrelevant news. The system can timely grab, reasonably classify, sort and analyze massive negative news by using AI technology, extracts the negative news which is considered by investors to be related to the main body default risk, and greatly improves the efficiency of review by users.
Application of stock market information
The market information of the stocks and the bonds is often related to the situation of the issuing main body and the related bonds, and the dynamic trend of the market reflects different risk levels. The system brings the securities market information into the securities market information, not only successfully realizes the prediction of default probability of an issuing subject in a default probability prediction model, but also assists an investor to insights the default condition of the bond by utilizing the credit profit difference analysis function of the bond.
Wherein the integrated financial rating:
a scoring card rating mechanism based on a Logistic model is brought into general industrial and commercial enterprises, city investment enterprises and real estate characteristic rating models, significance and relevance of related quantitative financial indexes of debt subjects to default are deeply researched, and credit rating assessment of the subjects is carried out for nearly 5 years. Based on the analysis of the basic surface of the main body, the result is often more solid and reliable, the coverage of the monitored object is wider, the credit rating distribution of the enterprises disclosed by the financial reports is effectively compared with the common virtual credit rating distribution of the external credit rating organization, the comprehensive financial rating of the system has the rating distribution which is consistent with the actual credit level of China, and the credibility is higher.
And (3) default probability prediction:
based on Merton option pricing as a theoretical basis, stock price quotations, financial reports and the like as data bases, a certain future deadline (generally 1 year) is given, the probability that the asset value of an enterprise is smaller than a default point is predicted, and then the default possibility of the enterprise is known. The system supports the analysis of the EDF value and the trend of an enterprise in nearly 8 years per day, and is expanded to the industry and regional analysis. The method can map the listed companies which stop the card for a long time and the non-listed companies which do not have stock price information to obtain results. Based on the stock price quotation, the daily risk monitoring result can be obtained, and the market change result can be reflected in time, so that the method has predictability and foresight.
The credit interest difference utilizes the bond market information, and the credit risk change of a main body corresponding to the bond can be reflected in time according to the bond credit interest difference change, so that a complementary prediction effect is achieved when the EDF fails to disclose default risks in advance.
Negative public opinion monitoring:
by taking artificial marking of more than 1 ten thousand news as the basis of public opinion analysis and combining natural language processing technologies such as semantic analysis and emotion classification, 10 ten thousand negative news in nearly one hundred days can be acquired later and extraction of more than 3000 keywords can be completed, the news attitudes can be intelligently sensed and the negative news information related to credit can be accurately positioned. And meanwhile, the extracted news information is classified and analyzed in seven subjects, so that the user is assisted to quickly master the summary of each news information. In addition, the capturing and analyzing of mass news are expanded to the level of industry and related parties, and investors are assisted to know the dynamic and indirect negative news of the macro environment.
The method can be used for acquiring the relevant negative news of the main body concerned by the user most timely, acquiring a first-hand information source to classify seven major subjects of the negative news, and assisting investors in knowing the type distribution and the dynamic development trend of the key negative news of the main body.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic diagram of the overall architecture of the composite enterprise credit warning system of the present patent.
Fig. 2 is a schematic view of an accumulated broken line of the negative public sentiment theme of the enterprise obtained by the comprehensive analysis of the negative public sentiment display system of the enterprise.
Fig. 3 is an enterprise negative public opinion theme radar chart obtained by comprehensive analysis of the enterprise negative public opinion display system.
FIG. 4 is a schematic diagram illustrating EDF value prediction of probability that the asset value of an enterprise is less than a default point.
In the attached fig. 2: 1. (ii) a capability; 2. will; 3. a condition; 4. compliance; 5. a market place; 6. a high pipe; 7. and others.
Detailed Description
The present invention will be described in further detail with reference to examples, which are illustrative of the present invention and are not to be construed as being limited thereto.
Example (b):
as shown in fig. 1, the present patent provides a composite enterprise credit early warning system, which includes a negative public opinion module, a default probability module, a financial rating module, and a comprehensive analysis module; the negative public opinion module comprises an enterprise public opinion information acquisition system, an enterprise negative public opinion processing system and an enterprise negative public opinion display system; the default probability module comprises an enterprise security information acquisition system, an enterprise default probability prediction system and a default probability result display system; the comprehensive financial module comprises an enterprise comprehensive financial information acquisition system, an enterprise comprehensive financial rating system and a financial rating result display system; and the comprehensive analysis module comprehensively analyzes the analysis results of the negative public opinion module, the default probability module and the financial rating module and then displays the whole result. The three modules of the system form an early warning mode with high timeliness and high coverage.
Table 1:
in a negative public opinion module of the composite enterprise credit early warning system, an enterprise public opinion information acquisition system comprises a target enterprise public opinion text acquisition unit through a crawler technology; the enterprise negative public opinion processing system comprises an enterprise public opinion text analysis unit and a processing unit; the enterprise negative public sentiment display system displays the enterprise negative public sentiments obtained by comprehensive analysis in a theme accumulation line graph (the line graph related to a specific enterprise is shown in detail in figure 2) or a theme radar graph (the radar graph related to the specific enterprise is shown in detail in figure 3). An enterprise public opinion text analysis unit in the composite enterprise credit early warning system performs natural language analysis on enterprise public opinion texts, including emotion classification technology, to obtain basic language features, wherein the basic language features are used for evaluating adverse effects on enterprise credit; inputting basic language features into a preset public opinion analysis and evaluation model to obtain a theme accumulation line graph or a theme radar graph of the negative public opinions of the enterprise, wherein the theme classification of the negative public opinions of the enterprise in the composite enterprise credit early warning system comprises the following steps: capabilities, willingness, status, compliance, market, high management, and other categories.
In a default probability module of the composite enterprise credit early warning system, the enterprise security information acquisition system comprises an acquisition unit for capturing target enterprise security information texts in security information libraries of various companies stored in the system, and the information libraries can also be directly connected with the existing security systems in the market; the enterprise default probability prediction system comprises an enterprise security information text analysis unit and a processing unit; and the default probability result display system displays the default probability result through the EDF value. The text analysis unit and the processing unit of the enterprise security information in the composite enterprise credit early warning system draw an asset value time curve of an enterprise through text analysis by taking the Merton option pricing as a theoretical basis, and predict a probability EDF value that the asset value of the enterprise is smaller than a default point by giving a certain future time limit on the basis of the stock price quotation and the financial newspaper text information captured in the data acquisition unit, as shown in detail in FIG. 4.
The enterprise financial information acquisition system comprises a system for acquiring a target enterprise comprehensive financial information text through a crawler technology; the enterprise comprehensive financial rating system comprises a text analysis unit and a processing unit; the results of the ranking of the corporate finances are shown by A +, A, A-and the revenue, net profit, and net cash flow for the business are shown by the bar graph. A text analysis unit and a processing unit of the enterprise comprehensive financial rating system score each financial index, index conversion and index weight of an enterprise by adopting a scoring card rating mechanism based on a Logistic model, then perform preliminary rating and model external adjustment to obtain final rating; the method comprises the steps of establishing a general industrial and commercial enterprise, a city investment enterprise and a real estate characteristic rating model, establishing the significance and the correlation coefficient of related quantitative financial indexes and default of a target enterprise, and evaluating the credit rating of the target enterprise in the last 5 years.
The enterprise financial information rating system comprises the following contents:
table 2:
the composite enterprise credit early warning system also comprises a target enterprise input system and a comprehensive collection early warning result display system; the target enterprise input system inputs a target enterprise to be inquired through setting a dialog box, and the target enterprise enters the composite enterprise credit early warning system for operation; the comprehensive collection early warning result display system comprehensively displays the results of the negative public opinion module, the default probability module and the comprehensive financial module through the table.
A composite enterprise credit early warning method comprises the following steps:
the method comprises the following steps: inputting a target enterprise to be searched through a target enterprise input system dialog box;
step two: triggering a negative public opinion module, a default probability module and a comprehensive financial module of the composite enterprise credit early warning system through the name of a target enterprise, wherein the enterprise public opinion information acquisition system captures enterprise public opinion information, the enterprise security information acquisition system captures security information texts of the enterprise, and the enterprise comprehensive financial information acquisition system captures comprehensive financial information texts of the enterprise;
step three: the enterprise negative public opinion processing system, the enterprise default probability prediction system and the enterprise comprehensive financial rating system respectively analyze and process the captured enterprise negative public opinion information text, the enterprise security information text and the enterprise comprehensive financial information text to obtain respective analysis results;
step four: and respectively displaying the results of each module in the third step through a result output unit, and performing table summarizing and displaying on the results of the three modules through a comprehensive collection early warning result display system.
In addition, it should be noted that the names and the like of the respective systems of the embodiments described in the present specification may be different. All equivalent or simple modifications made according to the principles described in the present patent concepts are included in the scope of protection of the present patent. Various modifications, additions and substitutions for the specific embodiments described may be made by those skilled in the art without departing from the scope of the invention as defined in the accompanying claims.
Claims (10)
1. The utility model provides a compound enterprise credit early warning system which characterized in that: the enterprise credit early warning system comprises a negative public opinion module, a default probability module, a financial rating module and a comprehensive analysis module;
the negative public opinion module comprises an enterprise public opinion information acquisition system, an enterprise negative public opinion processing system and an enterprise negative public opinion display system;
the default probability module comprises an enterprise security information acquisition system, an enterprise default probability prediction system and a default probability result display system;
the financial rating module comprises an enterprise comprehensive financial information acquisition system, an enterprise financial analysis system and a financial rating result display system;
and the comprehensive analysis module comprehensively analyzes the analysis results of the negative public opinion module, the default probability module and the financial rating module and then displays the whole result.
2. The composite enterprise credit warning system of claim 1, wherein: in the negative public opinion module, the enterprise public opinion information acquisition system comprises a target enterprise public opinion text acquisition unit through a crawler technology and a text interface;
the enterprise negative public opinion processing system comprises an enterprise public opinion text analysis unit and a processing unit; the enterprise negative public opinion display system displays the enterprise negative public opinions obtained through comprehensive analysis by a theme accumulated line graph or a theme radar graph.
3. The composite enterprise credit warning system of claim 2, wherein: the enterprise public opinion text analysis unit carries out natural language analysis on enterprise public opinion texts, wherein the natural language analysis comprises emotion classification technology, basic language features are obtained, and the basic language features are used for evaluating adverse effects on enterprise credit to obtain a theme accumulated broken line graph or a theme radar graph of enterprise negative public opinions.
4. The composite enterprise credit warning system of claim 2 or 3, wherein: the topic classification of the negative public sentiments of the enterprises comprises the following steps: capabilities, willingness, status, compliance, market, high management, and other categories.
5. The composite enterprise credit warning system of claim 1, wherein: in the default probability module, the enterprise security information acquisition system comprises an acquisition unit of target enterprise security information;
the enterprise default probability prediction system comprises an enterprise security information analysis unit and a processing unit; and the default probability result display system displays the default probability result through the EDF value.
6. The composite enterprise credit warning system of claim 4, wherein: the enterprise security information analysis unit and the processing unit take Merton option pricing as a theoretical basis, draw an asset value time curve of an enterprise through analysis, give a certain future time limit on the basis of stock price quotation and financial and newspaper information captured in the data acquisition unit, and predict the probability EDF value that the asset value of the enterprise is smaller than a default point.
7. The composite enterprise credit warning system of claim 1, wherein: in the financial rating module, the enterprise comprehensive financial information acquisition system comprises a text acquisition system for target enterprise comprehensive financial information through a crawler technology;
the enterprise comprehensive financial rating system comprises a text analysis unit and a processing unit;
the comprehensive financial rating result display system displays the rating result of the comprehensive financial of the enterprise through A +, A, A-and the operating income, net profit and net cash flow of the enterprise through a histogram.
8. The composite enterprise credit warning system of claim 7, wherein: the text analysis unit and the text processing unit score each financial index, index conversion and index weight of an enterprise by adopting a scoring card rating mechanism based on a Logistic model, then perform preliminary rating and model external adjustment to obtain final rating;
the enterprises comprise general industrial and commercial enterprises, city investment enterprises and real estate characteristic rating models, the significance and the correlation coefficient of related quantitative financial indexes and default of target enterprises are established, and the credit rating of the target enterprises in 5 years is evaluated.
9. The composite enterprise credit warning system of claim 1, wherein: the composite enterprise credit early warning system also comprises a target enterprise input system and a comprehensive collection early warning result display system;
the target enterprise input system inputs a target enterprise to be inquired through setting a dialog box, and enters the composite enterprise credit early warning system for operation;
the comprehensive collection early warning result display system comprehensively displays the results of the negative public opinion module, the default probability module and the comprehensive financial module through a table.
10. A composite enterprise credit early warning method is characterized in that: the method comprises the following steps:
the method comprises the following steps: inputting a target enterprise to be searched through a target enterprise input system dialog box;
step two: triggering a negative public opinion module, a default probability module and a financial rating module of the composite enterprise credit early warning system through the name of a target enterprise, wherein the enterprise public opinion information acquisition system captures enterprise public opinion information, the enterprise security information acquisition system captures security information texts of the enterprise, and the enterprise comprehensive financial information acquisition system captures comprehensive financial information texts of the enterprise;
step three: the enterprise negative public opinion processing system, the enterprise default probability prediction system and the enterprise comprehensive financial rating system respectively analyze and process the captured enterprise negative public opinion information text, the enterprise security information text and the enterprise comprehensive financial information text to obtain respective analysis results;
step four: respectively displaying the results of each module in the third step through a result output unit, and performing comprehensive early warning display on the results through a comprehensive negative public opinion module, a default probability module and a financial rating module; the display platform terminal equipment comprises a mobile phone, a tablet personal computer and a desktop computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010442991.4A CN113780604A (en) | 2020-05-22 | 2020-05-22 | Composite enterprise credit early warning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010442991.4A CN113780604A (en) | 2020-05-22 | 2020-05-22 | Composite enterprise credit early warning system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113780604A true CN113780604A (en) | 2021-12-10 |
Family
ID=78834161
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010442991.4A Pending CN113780604A (en) | 2020-05-22 | 2020-05-22 | Composite enterprise credit early warning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113780604A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803204A (en) * | 2017-01-17 | 2017-06-06 | 上海聚赟大数据有限公司 | A kind of floater's promise breaking analysis method based on machine learning |
| CN107527288A (en) * | 2017-06-16 | 2017-12-29 | 兴业数字金融服务（上海）股份有限公司 | A kind of bond survival phase Warning System |
| CN108932577A (en) * | 2018-04-25 | 2018-12-04 | 广州广电研究院有限公司 | A kind of assessment of business risk and early warning system |
| CN109165840A (en) * | 2018-08-20 | 2019-01-08 | 平安科技（深圳）有限公司 | Risk profile processing method, device, computer equipment and medium |
| CN110415125A (en) * | 2019-08-02 | 2019-11-05 | 浙江爱宽投资管理有限公司 | A kind of method, apparatus, equipment and computer storage medium carrying out credit rating to issuing subject online |
| CN110704572A (en) * | 2019-09-04 | 2020-01-17 | 北京航空航天大学 | Suspected illegal fundraising risk early warning method, device, equipment and storage medium |
| CN111105311A (en) * | 2019-12-19 | 2020-05-05 | 第一创业证券股份有限公司 | Dynamic credit rating method and device for bond body |
-
2020
- 2020-05-22 CN CN202010442991.4A patent/CN113780604A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803204A (en) * | 2017-01-17 | 2017-06-06 | 上海聚赟大数据有限公司 | A kind of floater's promise breaking analysis method based on machine learning |
| CN107527288A (en) * | 2017-06-16 | 2017-12-29 | 兴业数字金融服务（上海）股份有限公司 | A kind of bond survival phase Warning System |
| CN108932577A (en) * | 2018-04-25 | 2018-12-04 | 广州广电研究院有限公司 | A kind of assessment of business risk and early warning system |
| CN109165840A (en) * | 2018-08-20 | 2019-01-08 | 平安科技（深圳）有限公司 | Risk profile processing method, device, computer equipment and medium |
| CN110415125A (en) * | 2019-08-02 | 2019-11-05 | 浙江爱宽投资管理有限公司 | A kind of method, apparatus, equipment and computer storage medium carrying out credit rating to issuing subject online |
| CN110704572A (en) * | 2019-09-04 | 2020-01-17 | 北京航空航天大学 | Suspected illegal fundraising risk early warning method, device, equipment and storage medium |
| CN111105311A (en) * | 2019-12-19 | 2020-05-05 | 第一创业证券股份有限公司 | Dynamic credit rating method and device for bond body |
Non-Patent Citations (1)
| Title |
|---|
| 甄士龙;黎艳;: "用Logistic模型估计企业的违约概率", 广西金融研究, no. 11, pages 254 - 61 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Terblanche et al. | The influence of integrated reporting and internationalisation on intellectual capital disclosures | |
| Zamore et al. | Credit risk research: Review and agenda | |
| Liu et al. | Gone with the wind: An externality of earnings pressure | |
| CN108596439A (en) | A kind of the business risk prediction technique and system of knowledge based collection of illustrative plates | |
| US20130138577A1 (en) | Methods and systems for predicting market behavior based on news and sentiment analysis | |
| CN109191279A (en) | Medium-sized and small enterprises assessing credit risks platform based on supply chain finance on line | |
| CN105787073A (en) | Enterprise credit assessment method based on big data mining technology | |
| Huang et al. | Unsupervised neural networks approach for understanding fraudulent financial reporting | |
| CN113449964A (en) | Enterprise financial risk monitoring and early warning system and monitoring and early warning method | |
| CN110619462A (en) | Project quality assessment method based on AI model | |
| CN113988726A (en) | Enterprise industry credit evaluation management system based on block chain | |
| Zhang et al. | A study on SMIE credit evaluation model based on blockchain technology | |
| Zhu et al. | A new prediction NN framework design for individual stock based on the industry environment | |
| CN110222180A (en) | A kind of classification of text data and information mining method | |
| NGUYEN et al. | Factors affecting the implementation of environmental accounting by construction companies listed on the Ho Chi Minh Stock Exchange | |
| Mitrović et al. | Fraud and forensic accounting in the digital environment of accounting information systems: focus on the hotel industry | |
| CN111951105A (en) | Intelligent credit wind control system based on multidimensional big data analysis | |
| CN113780604A (en) | Composite enterprise credit early warning system and method | |
| CN114997588A (en) | Financial enterprise risk identification early warning method and system | |
| Wang | Early Warning and Monitoring Analysis of Financial Accounting Indicators of Listed Companies Based on Big Data | |
| Ding et al. | LSTM Deep Neural Network Based Power Data Credit Tagging Technology. | |
| Liu | Artificial Intelligence and Machine Learning based Financial Risk Network Assessment Model | |
| Li et al. | LSTM Deep Neural Network Based Power Data Credit Tagging Technology | |
| Burke et al. | Using a Large Language Model for Accounting Topic Classification | |
| Zhang | Application of Innovative Risk Early Warning Model Based on Big Data Technology in Internet Credit Financial Risk |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |