Method and device for obtaining specific user information

Abstract

The invention discloses a method and a device for acquiring information of specific users. The method comprises the following steps of acquiring operation behavior information, which is recorded in a system, of various first users; acquiring statistic data, which are recorded in the system, of various second users; establishing a specific second user gather according to values of the second users on various variables; judging whether the second users relevant to the first users in operation behaviors of the first users are specific second users or not according to the operation behavior information of the first users and the information of the specific second user gather; judging whether the first users are specific first users or not according to a judgment result and the operation behavior information of the first users; establishing a specific first user gather; and returning the information of the specific first user gather and / or the information of the specific second user gather to relevant application procedures. By the method and the device for acquiring the information of the specific users, the specific first user gather and the specific second user gather can be objectively and comprehensively established, and a data foundation is provided for applications or services such as information recommendation.

Description

technical field [0001] This application relates to the technical field of data mining, in particular to a method and device for obtaining specific user information. Background technique [0002] Computer applications or service systems generally have their own user groups, and generally hope to make full use of various information recorded in the system to improve or enrich system functions and other aspects. [0003] For example, in a third-party trading platform, there are a large number of seller users, and the amount of product information released is even greater. At this time, how to help buyer users find the product information they need more conveniently and quickly is the trading platform's own function. issues to be considered. For this reason, the trading platform can recommend information to users based on the similarity between buyer users. That is, for the current buyer user A, find N buyer users most similar to the buyer user in terms of shopping preferences...

AI Technical Summary

Problems solved by technology

[0004] However, due to the need to find N buyers most similar to the current buyer user from all buyer users on the trading platform, and the number of buyer users is very large, the calculation amount will be very huge, which seriously consumes system resources; on the other hand, eventually Information such as the quality of the product information in the store recommended to the buyer user is agnostic

In this way, even if the current buyer user is really interested in the recommended results, if the final purchased product has problems such as poor quality, it may enter the return process, etc. This is equivalent to the fact that the recommendation made by the trading platform is actually invalid , and will also reduce the buyer's user's trust in the trading platform

Images