Merchant evaluation method and system

1. A merchant evaluation method, comprising:

acquiring, by a processor, a merchant black sample and a merchant white sample of a plurality of merchants, wherein the merchant black sample includes a merchant receiving negative user feedback, and the merchant white sample includes a merchant receiving positive user feedback;

establishing, by the processor, a merchant evaluation model for evaluation indexes in multiple dimensions based on the merchant black sample and the merchant white sample, wherein the merchant evaluation model is established based on supervised regression modeling;

acquiring, by the processor, multi-dimensional evaluation index data of a first merchant to be evaluated, wherein the multi-dimensional evaluation index data comprises any combination of merchant background identity evaluation index data, merchant operation history evaluation index data, merchant operation capability evaluation index data, merchant business relationship evaluation index data, and merchant operation characteristics evaluation index data;

obtaining, by the processor, a health portrait and a health score of the first merchant to be evaluated using the merchant evaluation model, wherein obtaining the health portrait comprises: inputting the multi-dimensional evaluation index data into the merchant evaluation model; and obtaining, by the merchant evaluation model using calculation based on standard normal distribution, a probability value of evaluation index data in each dimension of the multiple dimensions, and wherein obtaining the health score comprises: determining a comprehensive score based on probability values of evaluation index data in all dimensions of the multiple dimensions, by calculating a geometric mean of the probability values of the evaluation index data in all dimensions of the multiple dimensions;

receiving, by the processor, a request for the health portrait and the health score of the first merchant for e-commerce shopping from a user; and

in response to the request, displaying, by the processor on a screen, the health portrait and the health score of the first merchant, wherein displaying the health portrait comprises presenting a chart with the evaluation indexes of the multiple dimensions associated with the merchant evaluation model and the probability value of the evaluation index data in each dimension of the multiple dimensions, and wherein displaying the health score comprises displaying the determined comprehensive score.

2. The method according to claim 1 , wherein establishing, by the processor, a merchant evaluation model comprises:

establishing a logistic regression model as the merchant evaluation model.

3. The method according to claim 1 , wherein obtaining the health portrait of the first merchant to be evaluated comprises:

synthesizing the probability value of the evaluation index data in each dimension to obtain the health portrait of the first merchant to be evaluated, wherein the health portrait of the first merchant to be evaluated reflects normal distribution of the first merchant to be evaluated in merchant evaluation historical data.

4. The method according to claim 1 , further comprising:

obtaining the evaluation index data in each dimension from original evaluation index description information, wherein the original evaluation index description information is inputted into an index model preset for an index to obtain evaluation index data used for describing a score of the index.

5. A merchant evaluation system, comprising:

a processor; and

a memory storing instructions that, when executed by the processor, cause the processor to perform:

acquiring a merchant black sample and a merchant white sample of a plurality of merchants, wherein the merchant black sample includes a merchant receiving negative user feedback, and the merchant white sample includes a merchant receiving positive user feedback;

establishing a merchant evaluation model for evaluation indexes in multiple dimensions based on the merchant black sample and the merchant white sample, wherein the merchant evaluation model is established based on supervised regression modeling;

acquiring multi-dimensional evaluation index data of a first merchant to be evaluated, wherein the multi-dimensional evaluation index data comprises any combination of merchant background identity evaluation index data, merchant operation history evaluation index data, merchant operation capability evaluation index data, merchant business relationship evaluation index data, and merchant operation characteristics evaluation index data;

obtaining a health portrait and a health score of the first merchant to be evaluated using the merchant evaluation model, wherein obtaining the health portrait comprises: inputting the multi-dimensional evaluation index data into the merchant evaluation model; and obtaining, by the merchant evaluation model using calculation based on standard normal distribution, a probability value of evaluation index data in each dimension of the multiple dimensions, and wherein obtaining the health score comprises: determining a comprehensive score based on probability values of evaluation index data in all dimensions of the multiple dimensions, by calculating a geometric mean of the probability values of the evaluation index data in all dimensions of the multiple dimensions;

receiving a request for the health portrait and the health score of the first merchant for e-commerce shopping from a user; and

in response to the request, displaying, on a screen, the health portrait and the health score of the first merchant, wherein displaying the health portrait comprises presenting a chart with the evaluation indexes of the multiple dimensions associated with the merchant evaluation model and the probability value of the evaluation index data in each dimension of the multiple dimensions, and wherein displaying the health score comprises displaying the determined comprehensive score.

6. The system according to claim 5 , wherein establishing the merchant evaluation model comprises:

establishing a logistic regression model as the merchant evaluation model.

7. The system according to claim 5 , the memory storing the instructions that, when executed by the processor, cause the processor to further perform:

synthesizing the probability value of the evaluation index data in each dimension to obtain the health portrait of the first merchant to be evaluated, wherein the health portrait of the first merchant to be evaluated reflects normal distribution of the first merchant to be evaluated in merchant evaluation historical data.

8. The system according to claim 5 , wherein the evaluation index data in each dimension is obtained from original evaluation index description information, wherein the original evaluation index description information is inputted into an index model preset for an index to obtain evaluation index data used for describing a score of the index.

9. A non-transitory computer-readable storage medium having stored therein instructions that, when executed by a processor of a device, cause the device to perform a merchant evaluation method, the method comprising:

acquiring a merchant black sample and a merchant white sample of a plurality of merchants, wherein the merchant black sample includes a merchant receiving negative user feedback, and the merchant white sample includes a merchant receiving positive user feedback;

establishing a merchant evaluation model for evaluation indexes in multiple dimensions based on the merchant black sample and the merchant white sample, wherein the merchant evaluation model is established based on supervised regression modeling;

acquiring multi-dimensional evaluation index data of a first merchant to be evaluated, wherein the multi-dimensional evaluation index data comprises any combination of merchant background identity evaluation index data, merchant operation history evaluation index data, merchant operation capability evaluation index data, merchant business relationship evaluation index data, and merchant operation characteristics evaluation index data;

obtaining a health portrait and a health score of the first merchant to be evaluated using the merchant evaluation model, wherein obtaining the health portrait comprises:

inputting the multi-dimensional evaluation index data into the merchant evaluation model; and obtaining, by the merchant evaluation model using calculation based on standard normal distribution, a probability value of evaluation index data in each dimension of the multiple dimensions, and wherein obtaining the health score comprises: determining a comprehensive score based on probability values of evaluation index data in all dimensions of the multiple dimensions, by calculating a geometric mean of the probability values of the evaluation index data in all dimensions of the multiple dimensions

receiving a request for the health portrait and the health score of the first merchant for e-commerce shopping from a user; and

in response to the request, displaying, on a screen, the health portrait and the health score of the first merchant, wherein displaying the health portrait comprises presenting a chart with the evaluation indexes of the multiple dimensions associated with the merchant evaluation model and the probability value of the evaluation index data in each dimension of the multiple dimensions, and wherein displaying the health score comprises displaying the determined comprehensive score.