Scenario-based query method and device, storage medium, and computer equipment

Abstract

The present application discloses a scene-based query method and device, storage medium, and computer equipment. The method includes: in response to a query request, obtaining a request vector corresponding to the query request, wherein the request vector includes a query feature vector , a user feature vector and a scene feature vector, the scene feature vector includes a position feature vector and / or a time feature vector; according to the matching degree between each feature vector included in the request vector and the store vector corresponding to the candidate store, the selected The target store in the candidate store. This application can effectively integrate scene features such as location and time, as well as query features and user features corresponding to query requests to recall stores, improve the recall effect of target stores, and help meet the personalized query needs of different local life users in different scenarios. This makes the store recall results more closely matched with the query scene and improves the user experience.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a scene-based query method and device, storage medium, and computer equipment. Background technique [0002] In the context of big data, how to provide users with personalized services based on user historical behavior data is an urgent problem to be solved in the implementation of big data applications. For the search recommendation system, recall is an important part, because the quality of recall directly affects the upper limit of the effect of subsequent steps. The recall stage mainly selects from massive information collections based on the characteristics of users and content (such as stores, products, advertisements, etc.). A part related to the current user preference is used as a candidate set, and then handed over to the sorting link. [0003] In the prior art, the recall stage generally calculates the matching degree between the user’s historical behavior...

AI Technical Summary

Problems solved by technology

However, in practical applications, this search recommendation method often leads to a high degree of similarity in the recommendation results of each search, that is, when users search for the same or similar keywords, the displayed results are roughly the same, and the recommendation effect is poor. Meet the diverse and personalized search needs of users in different situations, such as the need for different types of recommended meals at different times

Images