TWI719246B - Enterprise customer intelligent risk control system - Google Patents

Abstract

An enterprise customer intelligent risk control system includes a public news database, a negative rumors analysis server, a risk amount estimated database and a risk analysis electronic device. The public news database is connected to an external source of information to collect relevant news from an external website that has a creditor's relationship with a financial institution. The negative rumor analysis server is connected to the public news database and external sources to analyze the impact of the negative rumors based on the default model. The risk amount estimated database is connected to an internal source to calculate the amount of the risk based on the default. The risk analysis electronic device is connected to the negative rumor analysis server and the risk amount estimated database to calculate the risk value based on the default model.

Description

Enterprise customer intelligent risk control system

The invention relates to a financial credit management system, in particular to an intelligent risk control system for enterprise customers.

Credit management is a very important issue for every financial institution, especially when the financial institution has become the creditor of its corporate finance customers. How to make early warnings for corporate finance customers' negative news exposure due to credit degradation is an urgent problem to be solved at present.

The present invention proposes an innovative enterprise customer intelligent risk control system to solve the problems of the prior art.

In an embodiment of the present invention, an enterprise customer intelligent risk control system includes a corporate customer public news database, a negative rumors analysis server, a risk estimate database, and a risk control analysis electronic device. The corporate client's public news database is connected to an external data source, so as to collect relevant news on the external website that has a creditor's rights relationship with a financial institution according to preset conditions. The negative rumors analysis server is connected to the public news database of corporate customers and external data sources to analyze the impact of the relevant news of these corporate customers according to a preset model. The storm risk amount estimation database is connected to an internal data source to calculate the storm risk amount according to a preset method. The risk control analysis electronic device is connected to the negative rumors analysis server and the risk estimate database to calculate the risk value according to the preset model.

In an embodiment of the present invention, the corporate client public news database includes a communication interface for connecting to external data sources.

In an embodiment of the present invention, the corporate customer public news database further includes a processor connected to the communication interface for collecting relevant news of external websites according to the preset customer name.

In an embodiment of the present invention, the corporate customer public news database further includes a storage device connected to the processor to store the collected related news.

In one embodiment of the present invention, the negative rumor analysis server includes a communication interface connected to the public news database of corporate customers and external data sources.

In an embodiment of the present invention, the negative rumor analysis server further includes a processor connected to the communication interface to analyze the degree of influence of negative news based on keyword terms, Internet volume, and news concentration.

In an embodiment of the present invention, the risk estimate database includes a communication interface for connecting to an internal data source.

In an embodiment of the present invention, the risk risk estimate database further includes a processor connected to the communication interface for calculating the estimated risk risk range.

In an embodiment of the present invention, the risk control analysis electronic device includes a communication interface, which is connected to the negative rumors analysis server and the risk risk estimate database.

In an embodiment of the present invention, the risk control analysis electronic device further includes a processor connected to the communication interface for calculating the risk value according to the preset weight and grading according to the threshold.

In summary, the intelligent risk control system for enterprise customers of the present invention is to ensure the creditor's rights of financial institutions, collect public information on major social news and economic events, and build an intelligent analysis search system and a negative rumors analysis server. Using text analysis and artificial intelligence and other technologies, when major negative rumors occur to corporate customers of financial institutions, it automatically judges the early indicators of potential major events and finds out whether the debtor has harmed the bank’s claims. The system will automatically perform risk assessment and loss estimation After the production of warning information is confirmed by the relevant authorized personnel, the quota will be frozen and the transaction will be temporarily banned to ensure the creditor’s rights of financial institutions and control the risks of corporate financial customers.

Hereinafter, the above description will be described in detail by way of implementation, and a further explanation will be provided for the technical solution of the present invention.

In order to make the above and other objects, features, advantages and embodiments of the present invention more comprehensible, the description of the attached symbols is as follows:

100‧‧‧Intelligent Risk Control System for Enterprise Customers

110‧‧‧Electronic device for risk control analysis

112‧‧‧Communication interface

114‧‧‧Processor

116‧‧‧Storage Device

120‧‧‧The public news database of corporate clients

122‧‧‧Communication interface

124‧‧‧Processor

126‧‧‧Storage device

130‧‧‧Negative Rumor Analysis Server

132‧‧‧Communication interface

134‧‧‧Processor

136‧‧‧Storage Device

140‧‧‧Storm Risk Estimated Database

142‧‧‧Communication interface

144‧‧‧Processor

146‧‧‧Storage Device

150‧‧‧External data source

160‧‧‧Internal data source

170‧‧‧Mobile electronic device

In order to make the above and other objectives, features, advantages and embodiments of the present invention more comprehensible, the description of the accompanying drawings is as follows: Figure 1 is a block diagram of an enterprise customer intelligent risk control system according to an embodiment of the present invention Figure; and Figure 2 is a block diagram of an enterprise customer intelligent risk control system according to another embodiment of the present invention.

In order to make the description of the present invention more detailed and complete, please refer to the attached drawings and the various embodiments described below. The same numbers in the drawings represent the same or similar elements. On the other hand, well-known elements and steps are not described in the embodiments to avoid unnecessary limitations on the present invention.

In the implementation mode and the scope of the patent application, unless the article is specifically limited in the context, "一" and "the" can generally refer to a single or plural.

The technical aspect of the present invention is an intelligent risk control system for enterprise customers, which can be applied to various related operations requiring credit management. The following figures 1 and 2 are used to illustrate the specific implementation of the intelligent risk control system for enterprise customers.

Please refer to Figures 1 and 2. Figure 1 is a block diagram of an enterprise customer intelligent risk control system according to an embodiment of the present invention; Figure 2 is a block diagram of an enterprise customer intelligent risk control system according to another embodiment of the present invention Block diagram.

The enterprise customer intelligent risk control system 100 includes a risk control analysis electronic device 110, a corporate customer public news database 120, a negative rumors analysis server 130, and a storm risk amount estimation database 140. The corporate client public news database 120 and the negative rumor analysis server 130 are connected to an external data source 150 to continuously collect data according to preset conditions. The risk estimate database 140 is connected to an internal data source 160, which contains internal customer data or all business information of the financial institution. The external data source 150 is relative to the internal data source 160. Data that is not internal to a financial institution can be referred to as external data, such as public news information on external websites.

In one embodiment, the corporate customer public news database 120 is connected to an external data source 150, so as to collect news related to corporate customers that have a creditor's rights relationship with a financial institution on the external website according to preset conditions. For example, the corporate client public news database 120 can collect financial-related news (including positive or negative news) of corporate clients that have a creditor's rights relationship with the financial institution.

In one embodiment, the negative rumor analysis server 130 is connected to the corporate customer public news database 120 and the external data source 150 to analyze the influence of the related news of the corporate customers according to a preset model. For example, the negative rumor analysis server 130 has collected public news in the corporate client public news database 120, and only assesses the degree of influence of the negative news.

In one embodiment, the storm risk amount estimation database 140 is connected to the internal data source 160, so as to calculate the estimated storm risk amount range according to a preset method. For example, for corporate customers with suspected negative or credit depreciation signs, early risk assessment and loss estimation and notification and control operations are carried out. According to the relationship between the corporate customer (the debtor of the financial institution) and its related accounts, the customer's asset and liability information is aggregated, and after the aforementioned assets and liabilities are offset from each other, the customer's risk insurance amount can be evaluated and the scope of the estimated loss can be performed. Specifically, based on the relationship between the corporate customer (the debtor of the financial institution) and the related account, the data of the related account is retrieved. For example, companies with control and affiliation, companies with mutual investment relationships, companies whose chairman or general manager and the chairman or general manager of other companies are the same person or have a spouse relationship, interested parties and related credit guarantors . The above-mentioned guarantors include joint borrowers, joint invoicers, and bill debtors-refer to the guarantee endorsement of financial transactions, joint guarantors, collateral providers and joint guarantors, joint borrowers, joint debtors, collateral providers and joint debtors, and general guarantors , Collateral providers, etc.

In one embodiment, the risk control analysis electronic device 110 is connected to the negative rumors analysis server 130 and the risk estimate database 140 to calculate the risk value according to a preset model.

In one embodiment, the corporate client public news database 120 includes a communication interface 122 for connecting to an external data source 150. The communication interface 122 may be a network card of various protocol interfaces (for example, a wired or wireless network card).

In one embodiment, the corporate customer public news database 120 further includes a processor 124 connected to the communication interface 122 for collecting relevant news on external websites according to the preset customer name. For example, the corporate customer public news database 120 utilizes external website data retrieval management technology to extensively collect positive and negative reviews of corporate customers on the Internet, social groups, and electronic digital media, as well as major social news and economic event reports. From the traditional passive reception of major corporate customer rumors and information, it has been transformed into actively collecting comments on corporate financial customers in finance, social news and the community. Based on the collected data, target corporate customers (creditors of financial institutions) that are related to the creditor's rights of financial institutions. The so-called "customer name" can be the full name, abbreviation or code of an enterprise legal person.

In one embodiment, the corporate customer public news database 120 further includes a storage device 126 connected to the processor 124. The storage device 126 may be various forms of storage media, such as hard disk drive, solid state disk, floppy diskette drive, magnetic tape machine, etc., for storing The result of the execution of the above-mentioned processor 124, for example, relevant news collected.

In one embodiment, the negative rumor analysis server 130 includes a communication interface 132 for connecting to the corporate client public news database 120 and external data sources 150. The communication interface 132 may be a network card of various protocol interfaces (for example, a wired or wireless network card).

In one embodiment, the negative rumor analysis server 130 further includes a processor 134 connected to the communication interface 132 to analyze the degree of influence of negative news based on keyword terms, Internet volume, and news concentration. Negative rumors analysis server 130 uses text analysis and artificial intelligence technology to identify trends and associations hidden in the text information through 4W (When, What, Where, Who) and keyword and semantic analysis, and automatically judge the potential significance Advance indicators of events. Use public opinion analysis to find the "newest" and "most popular" news, and use the key indicators of positive and negative evaluation analysis: keyword terms, online voice volume, news concentration, as the criterion and basis for judging positive and negative rumors. Make important remarks that affect the credit rating of corporate finance customers. Then understand the spread and influence of major news on customers. Check whether the debtor endangers the bank's creditor's rights, and take preventive measures in advance. For "keywords", it refers to negative keywords, such as loan fraud, false accounting, default, capital reduction, stock price drop, etc. Regarding "Internet volume", it is based on the frequency of online news of the same event. The higher the volume of Internet volume, the higher the degree of influence. Regarding "news concentration", it is based on whether the same event has appeared in key/focus news or community discussions. If it has been exposed in key/focus news or community discussions, its influence will be higher. For example, in recent years, the media reported negative rumors of some corporate accounts, followed by the outbreak of loan fraud or abnormal credit granting problems (such as Dingxing loan fraud, Guangyang Technology fraudulent accounts, Phison's fake accounts, etc.). The system will screen out customers who are suspected of negative rumors (such as Dingxing, Guangyang, etc.) through the negative rumors preliminary indicators (negative keyword terms: fraudulent loans, false accounts, and online voice/news concentration: news report exposure/frequency) Branch, Group Alliance, etc.), obtain the client's overall compilation through the business directory, and perform follow-up risk assessment and loss estimation operations.

In one embodiment, the negative rumor analysis server 130 further includes a storage device 136 connected to the processor 134. The storage device 136 may be various forms of storage media, such as hard disk drive, solid state disk, floppy diskette drive, magnetic tape machine, etc., for storing The result of the execution of the aforementioned processor 134, such as the result of negative news and its analysis.

In one embodiment, the risk estimate database 140 includes a communication interface 142 for connecting to the internal data source 160. The communication interface 142 may be a network card of various protocol interfaces (for example, a wired or wireless network card).

In one embodiment, the risk risk estimate database 140 further includes a processor 144 connected to the communication interface 142 for calculating the estimated risk risk range. Based on the above-screened corporate customers (financial institution debtors) and their related customer data, retrieve and aggregate the customer's overall asset and liability information (including general deposits, accounts receivable, credit cards, investment securities, deposit pledges, etc.), After deducting the aforementioned assets and liabilities, the customer's risk insurance amount can be evaluated and the scope of loss can be estimated. For example, as shown in Table 1 and Table 2 below:

According to the relationship between corporate customers (financial institution debtors) and their related accounts in Tables 1 and 2 above, aggregate customer asset and liability information, and after deducting the aforementioned assets and liabilities from each other, you can evaluate the amount of customer risk and estimate the scope of loss. Therefore, the following calculation formula yields the risk insurance amount as: [(loan 45,123 thousand yuan)-deposit (25,002+100=25,102)] = 20,021 thousand yuan.

In one embodiment, the risk estimate database 140 further includes a storage device 146 connected to the processor 144. The storage device 146 may be various forms of storage media, such as hard disk drive, solid state disk, floppy diskette drive, magnetic tape machine, etc., for storing The result of the execution of the processor 144.

In one embodiment, the risk control analysis electronic device 110 includes a communication interface 112 for connecting to the negative rumors analysis server 130 and the risk estimate database 140 for data transmission.

In one embodiment, the risk control analysis electronic device 110 further includes a processor 114 connected to the communication interface 112 for calculating the risk score according to preset weights and grading according to the threshold. For example, the risk control analysis electronic device 110 can give scores and weights to the negative news influence and risk value estimation of corporate customers who have negative news, and then calculate the risk value and classify it as high, medium, or high according to subjective thresholds. Low-risk customers.

For corporate customers classified as high-risk, the system cannot automatically freeze the customer's available quota and temporarily ban transactions. Since the data sources of this system come from public information, there may be forged or deliberately spread false rumors in an attempt to influence the company's evaluation or stock price situation, so the accuracy of the data still requires manual intervention to confirm and verify. Therefore, the system automatically generates warning information (such as email and short message notification) and only sends it to the relevant authorized personnel for confirmation. If it does happen afterwards, for the high-risk corporate client, the quota will be frozen and the transaction temporarily banned, which will benefit the financial institution’s creditor’s rights assurance and corporate client risk control operations. This system can take early response measures and ensure the creditor's rights of financial institutions, so as to implement the principle of disclosure of internal risk information of financial institutions and strengthen the risk control and prevention mechanism.

The risk control analysis electronic device 110 further includes a storage device 116 connected to the processor 114 for storing the execution result of the processor 114.

In addition, when the risk control analysis electronic device 110 has determined that a corporate customer debtor has a high risk of harming bank claims, it will automatically notify the corporate customer debtor of the risk amount calculated by the aforementioned risk amount estimation database 140 Relevant authorized personnel confirm, and after confirmation, proceed with the processing of frozen quota and/or temporarily banned transactions. Regarding the notification method, for example, the risk control analysis electronic device 110 will automatically send a message to the mobile electronic device 170 of the relevant authorized personnel. The message can briefly describe the status of the enterprise customer, and attach the relevant link information for further evaluation and confirmation .

In summary, the intelligent risk control system for enterprise customers of the present invention is to ensure the creditor's rights of financial institutions, collect public information on major social news and economic events, and build an intelligent analysis search system and a negative rumors analysis server. Using text analysis and artificial intelligence and other technologies, when major negative rumors occur to corporate customers of financial institutions, it automatically judges the early indicators of potential major events and finds out whether the debtor has harmed the bank’s claims. The system will automatically perform risk assessment and loss estimation After the production of warning information is confirmed by the relevant authorized personnel, the quota will be frozen and the transaction will be temporarily banned to ensure the creditor’s rights of financial institutions and control the risks of corporate financial customers.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention. Anyone who is familiar with this technique can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection of the present invention The scope shall be subject to the definition of the attached patent application scope.

100‧‧‧Intelligent Risk Control System for Enterprise Customers

110‧‧‧Electronic device for risk control analysis

120‧‧‧The public news database of corporate clients

130‧‧‧Negative Rumor Analysis Server

140‧‧‧Storm Risk Estimated Database

150‧‧‧External data source

160‧‧‧Internal data source

170‧‧‧Mobile electronic device

Claims (6)

An intelligent risk control system for corporate customers, including: a corporate customer public news database connected to an external data source, so as to collect relevant news about corporate customers that have a creditor's rights relationship with a financial institution in an external website according to preset conditions; a negative The rumor analysis server is connected to the corporate client’s public news database and the external data source to analyze the impact of the corporate client’s related news based on a preset model. The negative rumor analysis server includes a processor and A communication interface, the communication interface is connected to the corporate client's public news database and the external data source, and the processor is connected to the communication interface to analyze the impact of negative news based on keyword terms, Internet volume and news concentration The negative rumors analysis server is used to use text analysis and artificial intelligence technology to find hidden trends and associations in text information through keyword and semantic analysis, automatically determine the early indicators of potential major events, and use public opinion Analyze and find the latest news and the hottest news, and use multiple key indicators of positive and negative evaluation analysis as the criterion and basis for judging positive and negative rumors to find out the important remarks that affect the credit rating of corporate finance customers. These key indicators Keyword terms, Internet volume and news concentration; a database of estimated risk amount, connected to an internal data source, to calculate the risk amount according to a preset method, among which, the estimated amount of risk amount data For one of these corporate customers that has been suspected of negative or degrading credit, the risk assessment and loss estimation and notification control operations are carried out in advance, and the risk assessment database is based on these corporate customers. The relationship between the customer and its related accounts, capture and consolidate the customer’s overall asset and liability information, and after the assets and liabilities are offset against each other, the risk assessment and the estimated loss range The account’s overall asset and liability information includes general deposits, accounts receivable, credit cards, investment securities, and deposit facilities; and a risk control analysis electronic device connected to the negative rumors analysis server and the risk amount estimate database , In order to calculate the risk value according to the preset model for the calculated amount of the risk, the risk control analysis electronic device includes a processor and a communication interface, and the communication interface of the risk control analysis electronic device is connected to the negative rumors The analysis server and the database of the estimated risk amount, the processor of the risk control analysis electronic device is connected to the communication interface of the risk control analysis electronic device to target the negative news impact of corporate customers who already have negative news Estimation of risk and risk amount are given scores and weights respectively, and then the risk value is calculated and classified into high, medium, and low-risk corporate customers based on subjective thresholds. According to claim 1, the enterprise customer intelligent risk control system, wherein the enterprise customer public news database includes: a communication interface for connecting to the external data source. For the enterprise customer intelligent risk control system described in claim 2, wherein the enterprise customer public news database further includes: a processor connected to the communication interface for collecting relevant news on external websites according to the preset customer name. For the enterprise customer intelligent risk control system described in claim 3, the enterprise customer's public news database further includes: a storage device connected to the processor to store the collected relevant news. The enterprise customer intelligent risk control system according to claim 1, wherein the risk estimate database includes: a communication interface for connecting to the internal data source. For the enterprise customer intelligent risk control system described in claim 5, the risk risk estimate database further includes: a processor connected to the communication interface for calculating the estimated risk risk range.

Images