Method for processing user credibility social network data

Abstract

A method for processing user credibility social network data provided by the invention is provided with a social network platform including WeChat, and includes: a step of user registration and generation of a corresponding social relation judging request; a step of obtaining different levels and types of social contact data network information while a user initiates a loan request; a step of data entry, cleaning and combing and output of user social contact data; a step of performing user credit assessment in a user social circle to obtain user credit data; and a step of performing secondary credit granting rating and a step of performing credibility assessment and judging the loan default rate and credit granting risk of the user. The method has the advantages that through estimation of the overdue rate of users, an assessed value of a model used for distinguishing user quality can reach above 40, an assessed value of a traditional model is only about 35, and the identification rate of user assessment can reach above 70%, i.e., repayment results of 70% of users can be accurately identified through the model. Remarkable effects are achieved in practical application compared with credit granting in a traditional mode.

Description

technical field

[0001] The invention relates to the field of Internet applications, in particular to a method for processing user credibility social network data. Background technique

[0002] At present, the product uses an online model to evaluate the quality of users, including but not limited to monitoring the authenticity, repayment ability and repayment willingness of users. Due to the fact that the authenticity of user data cannot be controlled when collecting data in the online mode, the user's repayment willingness and repayment ability cannot be evaluated and determined through the simple identity information and asset information uploaded by the user. At the same time, there is no sound user credit file data in China, and the vast majority of borrowers cannot query historical credit status through the credit interface, which makes it extremely difficult to evaluate the authenticity, repayment ability and repayment willingness of users online.

[0003] Facing that...

Examples

Embodiment 1

[0071] User A registers and enters the social platform through a mobile phone number and initiates a loan after completing identity authentication. The user implements the platform and requests to share the loan request to the social circle. A's friends A1, A2...A6 click the link, and record the click time and corresponding page refresh times of these 6 users. Then A1 and A2 among the 6 friends left the page, and the remaining 4 people A3, A4, A5, and A6 were guided to the next step by the page, that is, they answered the relevant questions about the identity of user A. A3 answered incorrectly, and A4 , A5, and A6 answered correctly and invested in user A. User A has raised a full amount and the data collection is complete.

[0072] For user A, based on the link A1-A6 has no abnormal values ​​in terms of time, region, and device matching, the value of the liked relationship is 6; based on the correct answers to A4-A6 and part A3, the value of the authentication relationship i...

Embodiment 2

[0078] User B registers and enters the social platform through a mobile phone number and initiates a loan after completing identity authentication. The user implements the platform's requirements and shares the loan request to the social circle. B's friends B1, B2, and B3 click the link, and record the click time and corresponding page refresh times of these three users. Then these three friends all entered the next step, that is, answered the relevant questions about the identity of the user, all answered correctly and invested in user B, user B raised a full amount, and the data collection was completed. For user B, based on B1-B3 clicking on the link, there are no abnormal values ​​in terms of time, region, and device matching, and the value of its like relationship is 3; based on the correct answer of B1-B3, its authentication relationship value is 3; based on B1-B3 investment If it is successful and no outliers appear, its investment relationship value is 3. At the same ...

technical field

[0001] The invention relates to the field of Internet applications, in particular to a method for processing user credibility social network data. Background technique

[0002] At present, the product uses an online model to evaluate the quality of users, including but not limited to monitoring the authenticity, repayment ability and repayment willingness of users. Due to the fact that the authenticity of user data cannot be controlled when collecting data in the online mode, the user's repayment willingness and repayment ability cannot be evaluated and determined through the simple identity information and asset information uploaded by the user. At the same time, there is no sound user credit file data in China, and the vast majority of borrowers cannot query historical credit status through the credit interface, which makes it extremely difficult to evaluate the authenticity, repayment ability and repayment willingness of users online.

[0003] Facing that...

Examples

Embodiment 1

[0071] User A registers and enters the social platform through a mobile phone number and initiates a loan after completing identity authentication. The user implements the platform and requests to share the loan request to the social circle. A's friends A1, A2...A6 click the link, and record the click time and corresponding page refresh times of these 6 users. Then A1 and A2 among the 6 friends left the page, and the remaining 4 people A3, A4, A5, and A6 were guided to the next step by the page, that is, they answered the relevant questions about the identity of user A. A3 answered incorrectly, and A4 , A5, and A6 answered correctly and invested in user A. User A has raised a full amount and the data collection is complete.

[0072] For user A, based on the link A1-A6 has no abnormal values ​​in terms of time, region, and device matching, the value of the liked relationship is 6; based on the correct answers to A4-A6 and part A3, the value of the authentication relationship i...

Embodiment 2

[0078] User B registers and enters the social platform through a mobile phone number and initiates a loan after completing identity authentication. The user implements the platform's requirements and shares the loan request to the social circle. B's friends B1, B2, and B3 click the link, and record the click time and corresponding page refresh times of these three users. Then these three friends all entered the next step, that is, answered the relevant questions about the identity of the user, all answered correctly and invested in user B, user B raised a full amount, and the data collection was completed. For user B, based on B1-B3 clicking on the link, there are no abnormal values ​​in terms of time, region, and device matching, and the value of its like relationship is 3; based on the correct answer of B1-B3, its authentication relationship value is 3; based on B1-B3 investment If it is successful and no outliers appear, its investment relationship value is 3. At the same ...

Images