Aggregation retrieval method and device based on multivariate data, equipment and storage medium

Abstract

The invention discloses an aggregation retrieval method and device based on multivariate data, equipment and a storage medium. The method comprises the steps of obtaining a to-be-retrieved target question, and performing intention recognition processing on the target question to obtain weight segmented words; using multiple retrieval engines to retrieve the weight segmented words, obtaining retrieval results corresponding to the retrieval engines, and the retrieval engines comprising a question and answer engine matched with the question and answer pair, a graph reasoning engine matched with the structured graph data and a universal retrieval engine matched with the unstructured document data; and performing confidence scoring on the retrieval results corresponding to the retrieval enginesaccording to the weights of the weight segmentation words by adopting an aggregation engine, and sequentially displaying the retrieval results of which the scores are greater than a preset thresholdvalue to the user. According to the technical scheme of the embodiment of the invention, aggregation retrieval is carried out on various types of data, the support range of retrieval data is expanded,and the retrieval accuracy is improved.

Description

Technical field[0001]The embodiments of the present invention relate to the field of computer technology, and in particular, to an aggregation retrieval method, device, equipment and storage medium based on multivariate data.Background technique[0002]With the continuous increase of the amount of information in various fields, the demand of users for information retrieval is also increasing. At present, most search engines are built based on traditional search technology, there is information overload, too many search results are returned, there are many search results that have nothing to do with the search intent, and there is a lack of accurate judgment of the user's actual search intent. The accuracy of retrieval is not high. At the same time, the search results are in a single form, and it is impossible to retrieve all the related data of multiple types and multiple structures in the vertical field, resulting in a higher frequency of secondary searches by users.Summary of the in...

AI Technical Summary

Problems solved by technology

At present, most search engines are built based on traditional search technologies, which have information overload, return too many search results, have many search results irrelevant to search intentions, and lack accurate judgments on users' actual search intentions. The search accuracy is not high

At the same time, the search results are in a single form, and it is impossible to retrieve relevant data of multiple types and structures in the vertical field, resulting in a high frequency of secondary searches by users

Images