TWM617622U - System for financial proof - Google Patents

Abstract

A system for financial proof is provided. The system provides an application platform that allows an applicant to submit application data for a specific item; an outbound system that provides an outbound interface for processing an outbound procedure for confirming the applicant’s working status; and a main server having a database. The method for financial proof is operated in the main server. Firstly, the main server receives the application data for an application item through the application platform. After the applicant is identified, the main server obtains the applicant’s identification data. The outbound system performs the outbound procedure to an external organization for confirming the applicant’s working status. Therefore, the system can rely on several financial characteristics such as applicant’s historical records, tax payment records and working status to determine whether or not the applicant is qualified to pass the application.

Description

Financial Proof System

The manual proposes a technology to prove financial strength, especially a system that can automatically perform financial proof based on the information and records provided by the customer.

If the general public has cash needs, such as loans, credit card applications, quota enhancements, etc., they need to prepare various financial proof documents, such as past credit information, financial information, payment records, or plus relevant information about the current company. Proof documents, so that financial units can conduct audits based on these financial proof documents, and if necessary, also conduct credit checks, and finally determine whether the applicant is eligible to apply through a loan or credit card.

However, according to the knowledge, the manual check of the applicant’s financial documents is often time-consuming, and it cannot solve the applicant’s urgent needs. There is an urgent need for a more efficient method.

In order to provide an effective method of reviewing financial resources, the disclosure document discloses a financial proof system in which computer technology is used to prove the applicant's financial resources quickly and efficiently through an automatically executed financial proof procedure.

According to an embodiment, a financial proof system is proposed. The system proposes an application platform that can provide applicants with application materials for an application project through a network interface; proposes an outbound system, provides an outbound interface, and can connect to a public switched telephone Network (PSTN), perform outbound procedures, including the ability to dial out to an external unit to confirm the applicant's working status; propose a main server with a database, which can be executed by one or more processors Proof of financial resources process.

In the process of implementing the financial proof, the application platform receives the application materials submitted by the applicant for an application project, and the identification data of the applicant can be obtained after the identity authentication of the main server. Then, based on the applicant's identification information, the applicant's identity information can be obtained, and then the outbound procedure can be executed through the outbound system to confirm the applicant's work status. Then you can obtain the applicant’s historical records or add tax payment records. According to one or more of the applicant’s work status, historical records and tax payment records, it can be determined whether the applicant’s financial resources are obtained Eligibility for the project.

Further, the outbound procedure implemented by the outbound system includes first obtaining the applicant’s working status from an external unit based on the applicant’s identification data, so that the repayment ability can be judged, and after obtaining the communication method of the external unit , Firstly form a voice message, and then convert the voice message to generate an outbound voice. After that, according to the communication method of the external unit, dial out to the external unit, and broadcast the outgoing voice after being connected. At this time, the external unit can use an interactive voice response system to respond to the voice asking for relevant information of the applicant. The voice message is also generated, received by the outbound system, and then recorded by the external unit. The external unit can be texted through the main server The spoken voice can interpret the applicant's working status from the textualized voice.

Further, in the above step of determining whether the applicant has obtained the qualification for the application project, the financial characteristics of the applicant are also evaluated by a grouping method. In this grouping method, first introduce a salary distribution model, which can be the result of salary distribution obtained from big data. Therefore, after comparing the salary distribution model, the group to which the applicant belongs can be judged based on the applicant’s identification data, and then The salary distribution of the applicant's group is determined, and the probability of the applicant passing the application project is judged.

Preferably, the application item proposed by the applicant is a loan or a line of credit, and the ranking is performed based on the credibility calculation of the financial characteristics, and whether the applicant is qualified for the application project is judged according to the highest ranking.

In order to further understand the features and technical content of the present invention, please refer to the following detailed descriptions and drawings about the present invention. However, the drawings provided are only for reference and explanation, and are not used to limit the present invention.

The following are specific specific examples to illustrate the implementation of this creation, and those skilled in the art can understand the advantages and effects of this creation from the content disclosed in this specification. This creation can be implemented or applied through other different specific embodiments, and various details in this specification can also be based on different viewpoints and applications, and various modifications and changes can be made without departing from the concept of this creation. In addition, the drawings of this creation are merely schematic illustrations, and are not depicted in actual size, and are stated in advance. The following implementations will further describe the related technical content of this creation in detail, but the disclosed content is not intended to limit the scope of protection of this creation.

It should be understood that although terms such as "first", "second", and "third" may be used herein to describe various elements or signals, these elements or signals should not be limited by these terms. These terms are mainly used to distinguish one element from another, or one signal from another signal. In addition, the term "or" used in this document may include any one or a combination of more of the associated listed items depending on the actual situation.

The disclosure discloses a financial proof system. First, please refer to the schematic diagram of the embodiment of the financial proof system structure shown in Figure 1. The system can be a computer host, a server, a workstation, a database, and a computer such as software programs running on each host. Technology achieved.

The figure shows that the system side is equipped with main components such as the application platform 12, the main server 14, and the outbound system 16, which can be connected to each other by means of network, line, etc., and can be realized by software with a computer host. The system submits an application platform 12, The application platform 12 can be implemented by a web server running on a specific server, and applications can be provided through various fields and forms on a network interface (such as a web page, mobile device software program (APP), etc.) A person submits application data for an application project through the Internet 10, including identification data related to the applicant's identity. The diagram shows that the applicant can use various user devices 101, 102, 103 to log in to the application platform 12 via the Internet 10. After identity authentication, the relevant application form can be obtained online, and the applicant can be guided to apply for and submit identity documents .

The system also proposes a dialing system 16. The dialing system can be a stand-alone computer host or server, or a software service running on a specific server. It provides an dialing interface to connect to the public switched telephone network (Public Switched Telephone Network). Telephone Network, PSTN) 18 or a specific network system. In this example, you can connect to an external unit 19 through the public switched telephone network 18, and perform an outbound dialing procedure through voice communication technology to dial out to the external unit 19 (such as labor insurance bureaus, insurance companies, telecommunications companies, relevant government departments, etc.) , The main purpose is to confirm the applicant’s working status, obtain the applicant’s work status and labor health insurance information such as on-job, resignation, unemployed, etc., and determine whether the applicant has the ability to repay, and the amount of repayment that can be afforded, etc. News.

The financial proof system also has a main server 14 that mainly runs financial proof methods, which has a database, such as a user database for managing user data, transaction history records of each user in a specific financial institution, or tax payment Records etc. The main server 14 can be implemented by a computer host, a server, or a computer system of a specific form, and one or more processors are used to perform the financial proof method.

According to one embodiment, the applicant submits an application for specific financial services through the application platform 12, such as loans, credit card applications, quota enhancements and other financial services that require financial support. According to various materials submitted by the applicant, the financial certification system can obtain the applicant To facilitate subsequent use of user identification data for various data acquisition and verification.

Figure 2 then shows a schematic diagram of an embodiment of the functional module in the financial proof system.

According to the illustrated embodiment, the main server 14 is equipped with a database, which can be divided into a user database 201 for storing customer data according to its function. The basic information of each user (maybe an applicant for financial services) is recorded. Information, authentication information and communication information, such as user name, age, ID number, insurance information, contact information, work location, job title and telephone number, etc., and can be updated at any time. The database also includes a historical database 204, which records the transaction history data of each user in a specific financial institution, such as past credit information, loan information, credit card repayment records, etc., which can be one of the personal financial characteristics. The main server 14 is provided with a question bank 202. The question bank 202 records the voice information required to be dialed out to the external unit 19 according to various purposes, which can be combined to become the dial out voice.

The main server 14 is provided with a calculation module 203 for running the financial proof method. The calculation module 203 uses one or more processors and memory of the computer host to run the financial proof method, wherein the processor executes the method stored in the memory Code to perform automatic procedures for proving financial resources. The main server 14 can implement an application platform (12, Fig. 1) through the software program, or another computer host or server implements the application platform, so that the main server 14 can receive the applicant through the application platform (12) through the application platform (12). The application materials for specific application items can obtain the identification materials of the applicant after identity authentication.

The outbound system 16 of the financial proof system can obtain the user identification data 211 from the autonomous server 14. In this example, it connects to the public switched telephone network 18 through the outbound interface 212, and when dialing external units 19, they can communicate with each other by voice messages. According to the embodiment, the autonomous server 14 of the outbound system 16 obtains the applicant’s identification data, which records the applicant’s name, name, birthday, ID number and other data, and works with the question bank 202 to form a voice dialed to an external unit 19 The voice message is continuously converted into the outbound voice through the voice conversion module 213 and the natural voice model 216 of the outbound system 16 to be dialed out to the external unit 19 through the outbound interface 212 after dialing.

After that, according to the embodiment, the external unit 19 may manually or an interactive voice processing module (Interactive Voice Response, IVR) 221 automatically respond to the voice sent by the outbound system 16, and the voice receiving module 214 in the outbound system 16 That is, the voice generated by the external unit 19 is received by the voice processing module 215 and can be combined with the natural voice model 216 to be processed into text content. The text content may involve the applicant's job status, job title, employment company, insurance status, etc. The main server 14 can then be provided as one of the financial characteristics for judging whether the applicant has passed the application qualification.

For example, if the external unit 19 is a joint solicitation center, the outbound system 16 can confirm the latest company based on the applicant's identity information. The external unit 19 can be a public data server that provides inquiries about specific company information, uniform numbers, scale, operating conditions and other information. After voice inquiries, it can respond to the applicant’s company information. Similarly, the response results can form the applicant’s One of the financial characteristics.

Refer to the flow chart of the main embodiment shown in FIG. 3 for the operation method of the relevant financial proof system.

In this process, at the beginning, as in step S301, the applicant fills in the application form through the web interface provided by the application platform, and the main server receives the applicant's application information for a certain application project through the application platform, and can obtain the applicant's identification Data, perform identity authentication, as in step S303. The method of identity authentication is not limited to a specific method. It can be authenticated through the applicant's image, identification data and application documents, or with other technologies (such as mobile devices, APP, email, etc.).

After that, when the outbound system obtains the user identification data, it can derive the appropriate one or more external units and corresponding communication methods (such as network communication methods, telephone numbers, etc.) based on the information therein, as in step S305, One of the purposes of the outbound procedure is to be able to obtain the applicant’s working status, including whether the applicant is employed, the company’s seniority and title, the applicant’s insurance information, etc., and the company information can also be obtained through a specific external unit, with a unified number, With reference to the general salary level of the relevant industry under its professional title, such information as the scale and operating conditions can be used to estimate the salary of the applicant, which becomes one of the applicant's financial characteristics, and then judges its repayment ability. For further outbound procedures, please refer to the flow of the embodiment shown in FIG. 4.

In addition to obtaining the above information from external organizations, in step S307, historical records can be obtained in the database of the main server system, or in step S309, the applicant's tax payment records can also be obtained through internal or external databases. Up to this step, the financial proof system can judge whether the applicant's financial resources are eligible for the application project based on one or more of the applicant's work status, historical records, and tax payment records.

However, regardless of whether the system can query the information sufficient to prove the applicant's financial resources, or whether the applicant can submit any documents that can prove his financial resources, the system can still continue to execute the clustering method, such as step S311, the clustering method can apply a clustering algorithm Method (such as K-means), which is a method of grouping applicants according to their name, age, industry, title and seniority of the company based on the results of data analysis. The results of the grouping can estimate the applicant’s salary (weekly salary, Monthly salary, annual salary, etc.), and evaluate whether there is a chance that the financial conditions for obtaining a loan cannot be met? If the chance is not passed, another classification algorithm (such as Light GBM) is performed to determine whether the applicant's salary meets the financial conditions (lower salary limit) for obtaining a loan. For the grouping method, refer to the embodiment process shown in FIG. 5.

The financial proof method operated by the disclosure proposal system is to obtain the applicant's historical record in step S307, step S309 to obtain the applicant's tax payment record, and/or the result of grouping in step S311 (one or any combination). One or more financial characteristics allow the system to determine whether the applicant’s financial resources are eligible for the project.

According to an embodiment, the above financial characteristics can be a credibility as a criterion for judgment, and the calculation method of the credibility is mainly to evaluate the weights based on the sources of acquisition of various financial characteristics. For example, if the applicant’s tax payment records are derived from official data, the weight of credibility is relatively high; if the applicant’s transaction records are scattered across different sources (such as banks), the transaction records obtained by the system will not Incomplete, the credibility of this financial feature is poor, and the weight assigned to this historical record is not high; if the result obtained by the applicant in the clustering method only just crosses the lower limit of salary, the weight given is not high. By analogy, the system can perform credibility evaluations based on various financial characteristics, and then, in step S313, perform a credibility ranking of the above historical records, tax payment records, and group classification results, so that the system can perform credibility rankings according to credibility. Whether the application items submitted by the applicant are passed, for example, in the credibility ranking, judge the applicant's historical records, tax payment records, and group classification results with the highest credibility to determine whether the applicant is eligible for the application item , As in step S315 in the process.

Figure 4 shows a flow chart of an embodiment of the outbound procedure in the outbound procedure in the outbound procedure run by the method for the outbound procedure. The procedure is mainly used in the outbound system. The outbound system can be an independent host or server, or run by a software program in a specific server. For examples of services, please refer to the functional modules shown in Figure 2.

According to the embodiment of the outbound procedure executed by the outbound system, initially, in step S401, the outbound system receives the identification data of the user (applicant) provided by the main server, and in step S403, the outbound system The software program obtains the applicant's name, address, ID number and other identifiable information based on the identification data, and after querying the communication method of the external unit, it can generate voice information based on the question bank in the outbound system.

Then, in step S405, the software program in the outbound system is used to convert the voice information into the outbound voice, in step S407, the outbound voice is formed, and then in step S409, the outbound system executes outbound calls according to the communication mode of the external unit , Such as dialing a phone number or using the Internet to communicate. Then, in step S411, the outgoing voice is played after the connection is made. At this time, in addition to manual voice communication, it can also be docked with the interactive voice response system in the external unit. The interactive voice response system can automatically respond to the voice issued by the outbound system. After that, in step S413, the software program in the outbound system receives and records the voice of the external unit, and in step S415, it is textualized to form a record, and the working status of the applicant can be interpreted through the textualized voice. Then, in step S417, it is stored in the outbound system. Finally, the result can be sent back to the main server. In step S419, the main server can obtain the applicant's working status based on the voice of the external unit provided by the outbound system.

According to the above embodiments, if in a situation, for example, when there is no (or incomplete, unreliable) history record or tax payment record, the above outbound procedure can be run to confirm the applicant's work status. In addition, the machine learning method can also be used to judge based on the information provided by the applicant (basic information, industry, marital status, job title, seniority, etc.) and joint information (joint credit agency), using the grouping method to determine the salary distribution based on the group Applicant's grouping results, and subsequent classification. The classification method is a technology based on one or more independent variables to determine the category of the strain number. It is applied to financial proof. Logistic regression can be used to establish questions for specific application items (which can be obtained from the question bank) to inquire about the application People have various questions, and the responses are classified into two categories, and the probability of meeting the eligibility of the application is calculated.

Figure 5 shows a flowchart of an embodiment of the grouping method in the financial proof method.

The main server starts to execute the grouping algorithm (step S501), and first introduces the salary distribution model (step S503). The embodiment can refer to the box-and-whisker diagram shown in FIG. The confirmed work status (such as whether they are employed, job title, seniority, industry category, etc.) determine the applicant's group (step S505), and obtain the probability of passing the application from the salary distribution model to determine whether there is a probability of failing the application (Step S507). Among them, according to the salary distribution in the salary distribution model (for example, refer to the upper and lower limits), when the applicant has no chance of failing to pass the application project (No), step S509 can be executed, and the system will confirm that the applicant is qualified, and further draw the application The credibility of the person passing the application item; on the contrary, when the applicant has a probability of not passing the application (Yes), as in step S511, the classification algorithm is still executed.

When the applicant has the possibility of failing to pass the application project, continue to implement a classification method. According to an embodiment, according to multiple characteristics of the applicant in the applicant’s identification data, a decision tree algorithm based on a histogram is adopted. , Find the best division point (decision tree) for classification (step S513), and finally evaluate the credibility of the applicant's financial characteristics according to the classification result to determine whether the application can be passed (step S515).

In this way, according to the description of the above embodiment, the financial proof system obtains the applicant’s historical records or adds tax payment records based on the applicant’s identification data, and then cooperates with the grouping or classification results to form one or more financial characteristics, and judges the application based on this Whether the person’s financial resources are eligible for the project.

Fig. 6 schematically shows an example of using box-and-whisker plots in the grouping algorithm.

This example shows the salary distribution model used by the grouping algorithm, showing the salary distribution in different ages (or other conditions), which means that the system can first group according to the salary status of each age group (or other conditions), and then judge the application The group of people, and can estimate the salary level of the applicant based on this. In this example, the box-and-whisker diagram shows the probability of salary distribution of each group of people, and the salary difference between each group is small. Each group represents the upper and lower limits of the group's salary, forming a group lower bound 601. After grouping, compare the lower group 601. If the applicant’s salary is not lower than the lower group 601, it means that the applicant’s salary exceeds the lower group 601, which means that the applicant can be qualified for the application; otherwise, if the applicant belongs to the group If the salary lower limit is lower than the group lower limit 601, the classification algorithm will continue to be executed.

In summary, according to the financial proof system disclosed in the above embodiments, the system automatically judges the financial resources of applicants applying for a certain project through data collection, big data analysis, machine learning, outsourcing procedures, clustering or classification methods, etc. In an application, the application item submitted by the applicant can be a loan or credit limit (such as credit card application, limit increase, etc.), and a ranking is made based on the credibility calculation of one or more of the above financial characteristics, according to the highest ranking among them Determine whether the applicant meets the qualifications of the application project.

The content disclosed above is only a preferred and feasible embodiment of the present model, and does not limit the scope of the patent application of the present model. Therefore, all equivalent technical changes made by using the description and schematic content of the present model are included in the application of the present model. Within the scope of the patent.

101, 102, 103: user device

10: Internet

12: Application platform

14: main server

16: Outbound system

18: Public Switched Telephone Network

19: External unit

201: User database

202: Question Bank

203: Calculation Module

204: Historical Database

211: User identification data

212: Outbound interface

213: Voice conversion module

214: Voice receiving module

215: Voice Processing Module

216: Natural Speech Model

221: Interactive voice processing module

601: The Lower Bound of the Group

Steps S301～S315: flow chart of financial proof method

Steps S401～S419: outbound program diagram

Steps S501～S515: flow chart of grouping method

Figure 1 shows a schematic diagram of an embodiment of the financial proof system architecture;

Figure 2 shows a schematic diagram of an embodiment of a functional module in the financial proof system;

Figure 3 shows a flow chart of a main embodiment of the financial proof system for executing financial proof;

Figure 4 shows a flowchart of an embodiment of the outbound program running in the financial proof system;

Figure 5 shows a flowchart of an embodiment of the grouping method in the financial proof system; and

Figure 6 shows an example schematic diagram of box-and-whisker plots used in the grouping algorithm.

101, 102, 103: user device

10: Internet

12: Application platform

14: main server

16: Outbound system

18: Public Switched Telephone Network

19: External unit

Claims (10)

A financial proof system includes: an outbound system, providing an outbound interface to execute an outbound procedure, and dial out to an external unit to confirm the working status of an applicant who submits an application item; and a main server, Including a user database storing user data and a historical database recording each user's transaction history data in a specific financial institution, connected to the outbound system to provide the outbound system to obtain the identification data of the applicant , Execute a financial proof method through one or more processors of the main server, which includes: receiving the applicant’s application data for the application item, after identity authentication, obtaining the applicant’s identification data; according to the application Person’s identification data, execute the outbound procedure through the outbound system to confirm the applicant’s work status; obtain the applicant’s historical records from the historical database based on the applicant’s identification data or add tax payment records ; And according to one or more of the applicant's work status, historical records and tax payment records, determine whether the applicant is eligible for the application project. The financial proof system as described in claim 1 further includes an application platform, which provides the applicant's application materials for the application project through a network interface. The financial proof system according to claim 2, wherein the application platform is a web server running on the main server or another server, which implements the network interface and is used to authenticate the identity of the applicant. The financial proof system as described in claim 1, wherein the outbound system is an independent computer host or server, and the outbound interface is connected to a public switched telephone network or a network system to execute the outbound procedure . The financial proof system according to claim 4, wherein the outbound procedure executed by the outbound system includes: forming a voice message based on the identification data of the applicant and the communication mode of the external unit; converting the voice message to generate a voice message Outbound voice; dial out to the external unit according to the communication method of the external unit, and broadcast the outbound voice after being connected; receive and record the voice from the external unit; and textualize the voice from the external unit, where , Interpret the applicant’s working status from the textual voice. According to claim 5, the financial proof system, wherein the main server is provided with a question bank, which records the voice information required for dialing out to the external unit to form the voice information according to various purposes, which can be combined to become the dialing voice. The financial proof system according to claim 6, wherein the voice message is converted into the outbound voice through a voice conversion module and a natural voice model of the outbound system, so as to be dialed out after dialing through the outbound interface To the external unit. The financial proof system according to claim 1, wherein in the step of determining whether the applicant has obtained the qualification of the application project, the financial characteristics of the applicant are also evaluated by a clustering method, and the clustering executed by the main server The method includes: introducing a salary distribution model; comparing with the salary distribution model, judging the group to which the applicant belongs based on the applicant’s identification data; and judging the applicant through the salary distribution of the group to which the applicant belongs. The probability of applying for the project. The financial proof system described in claim 8, wherein in the grouping method executed by the main server, when the applicant has no chance of failing the application item, the salary of the applicant is directly estimated and becomes Financial characteristics of the applicant One; when the applicant has a probability of failing to pass the application item, continue to implement a classification method, according to the applicant’s identification data, based on multiple characteristics of the applicant, through the decision tree algorithm based on histogram , To obtain the financial characteristics of the applicant to determine whether the application can be passed. For example, the financial proof system according to any one of claims 1 to 9, wherein the application item submitted by the applicant is a loan or a line of credit, and a ranking is performed based on the credibility calculation of the one or more financial characteristics, According to the highest ranking among them, judge whether the applicant meets the eligibility of the application project.

Images