Research Article
[Retracted] A Relevant Customer Identification Algorithm Based on the Internet Financial Platform
| | Input: Determination of the predictive result set A, user pairing set B, seed user relationship matrix, candidate set matrix; | | | Output: Best Paired Combination in Collection B; | | | The specific steps are as follows: | | (1) | Extract the pairing user in B, divide users into and ; | | (2) | Search all direct neighbors of in , labeled, deposits set ; | | (3) | Search all direct neighbors of in , labeled, deposits set ; | | (4) | Use the user in and to combine the collection , find a common neighbor of and , and label; | | (5) | The similarity of and is calculated by the following formula; | | (6) | The absolute difference is calculated by the following formula; | | (7) | Repeat the above steps until all the pairing users in the collection B are complete; | | (8) | Take the value to the user output, the pairing user corresponding to . |
|