Abnormal reason analysis method, device and server

Abstract

A method, an apparatus, a server and a storage medium for analyzing the cause of an abnormality are provided. The method includes determining a target recommendation result in which an abnormality exist in a recommendation system; acquiring a target request responded to when the recommendation system generates the target recommendation result; from the historical execution information for different requests, querying target execution information generated when the recommendation system generates the target recommendation result in response to the target request, and generating execution information corresponding to the recommendation result generated when the recommendation system generates the request in response to the request before the current time for the history execution informationof one request; analyzing a reason why the recommendation system generates the target recommendation result based on the target execution information, and obtaining an abnormal reason of the recommendation system. The present application analyzes and obtains the abnormal reason of the recommendation system on the basis of the target recommendation result generated by the real reproduction recommendation system, and improves the accuracy of the analyzed abnormal reason of the recommendation system.

Description

technical field [0001] The present application relates to the technical field of recommendation systems, in particular to a method, device and server for analyzing abnormal causes. Background technique [0002] Although the emergence and popularization of the Internet has brought a large amount of information to users and met the needs of users for information in the information age, there is usually a problem that users cannot quickly obtain the information they are interested in when faced with a large amount of information. In order to solve this problem, a recommendation system that can recommend information of interest to users according to their interest characteristics and behavior habits has emerged as the times require. [0003] In order to ensure the accuracy of the recommendation results generated by the recommendation system, it is usually necessary to analyze the reason why the recommendation system generates the recommendation results when there is an abnormali...

AI Technical Summary

Problems solved by technology

However, because the scenario of simulating access to the recommendation system by constructing a diagnostic analysis request is not a scenario where a real recommendation system generates abnormal recommendation results, the existing technology often analyzes the abnormality of the recommendation system by constructing a diagnostic analysis request. The abnormal reason of the recommendation system is inaccurate

Images