Information recommendation method and device, storage medium and electronic equipment

Abstract

The invention discloses an information recommendation method and device, a storage medium and electronic equipment, and can recall candidate commodities and determine historical behavior records of a user based on an information recommendation request sent by a user terminal. And then, for each historical behavior record, according to the environment information when the user executes the historical behavior record and the current environment information, determining an environment association weight corresponding to the historical behavior record, and according to the commodity feature of the commodity corresponding to each historical behavior record and the corresponding environment association weight, determining the interest feature of the user. And finally, according to the interest features of the user and the commodity features of the candidate commodities, predicting the click rate of the user to the candidate commodities, and performing information recommendation based on the click rate. By introducing the attention mechanism of the environmental information, the interest characteristics of the user are concentrated on the historical behavior record with relatively strong environmental association, so that the interest characteristics of the user can be represented more accurately, the prediction result is more accurate, and the recommendation hit rate is higher.

Description

technical field [0001] The present application relates to the technical field of information recommendation, and in particular to an information recommendation method, device, storage medium and electronic equipment. Background technique [0002] With the development of big data technology, more and more e-commerce platforms recommend products to users through recommendation systems to improve users' decision-making experience. [0003] At present, when recommending products to users, it is usually based on the list of products browsed, clicked or purchased by the user to predict the click rate of the user to be recommended products, so as to sort the products according to the predicted click rate. Specifically, the historical behavior records of the user may be obtained first, wherein the historical behavior records include the commodities browsed, clicked or purchased by the user. Secondly, input the product information of each product in the user's historical behavior re...

AI Technical Summary

Problems solved by technology

[0004] However, in a location-based service (Location Based Services, LBS) scenario, such as a takeaway scenario, user behavior records may be distributed in multiple geographical locations, and the products recalled by the recommendation system are limited by LBS, and can only recall the user's current location nearby products

At this time, based on the behavior records of users in different geographical locations, the click rate prediction for the currently recalled products is often low in prediction accuracy, resulting in a low hit rate for information recommendation

Images