Interaction event-based webpage project recommendation method and device

Abstract

The embodiment of the invention provides an interaction event-based webpage project recommendation method and device. The method comprises the steps of generating the relevance between interaction events and target interaction events in the interaction events according to interaction information of the interaction events of multiple users for webpage projects; for the interaction events of the current user, extracting a target webpage project corresponding to at least one target interaction event according to the relevance between the interaction events and multiple target interaction events; and recommending the target webpage project to the current user, so that the problem that effective webpage project recommendation cannot be carried out due to the fact that the users do not provide evaluation points is solved; furthermore, the webpage projects conforming to the user interests can be objectively and accurately recommended to the users by adopting the interaction events of the webpage projects as a basis of relevance computation; and only the relevance between the target interaction events and the interaction events is calculated in the relevance computation, so that the data quantity in the relevance computation is reduced and computing resources and storage resources of a server are saved.

Description

technical field [0001] The present application relates to the field of Internet technology, in particular to a method for recommending webpage items based on interactive events and a device for recommending webpage items based on interactive events. Background technique [0002] With the continuous development of Internet technology, more and more users browse various commodities on the web. [0003] In order to recommend webpage items that users may be interested in, the website provider will collect a large number of users' evaluation scores for each commodity, and use the evaluation scores to calculate the correlation matrix between commodities through a collaborative filtering algorithm. For a certain user, use the user's evaluation points for some products and the correlation between this part of products and other products to calculate the recommendation degree of other products and recommend them to the user accordingly. [0004] At present, there are three problems ...

AI Technical Summary

Problems solved by technology

[0004] At present, there are three problems in this product recommendation method: first, in practical applications, users may not want to spend time and effort to provide product evaluation points, resulting in the lack of evaluation points and cannot make effective product recommendations; The evaluation score has a subjective tendency, and the product recommendation based on the subjective evaluation score cannot objectively and accurately recommend the product that meets the user's interest to the user; finally, as the data volume of the product in the Internet continues to increase, through the collaborative filtering algorithm Commodity recommendation will rapidly expand the data volume of the correlation matrix in the current algorithm, resulting in overstorage of the server, overloading of computing and processing work, and even server crashes

Images