Entity relationship extraction method based on multi-target fusion

Abstract

The invention relates to an entity relationship extraction method based on multi-target fusion, belongs to the technical field of natural language processing, and aims to effectively solve the technical problems of inconsistent user cognition, low user portrait discovery and matching accuracy and the like caused by different message barriers of social APPs in the Internet. According to the method, chats, dynamic records and the like in different social APPs are extracted in combination with a deep learning feature extraction model and CRF decoding. Through constructing a record aggregation method, discovery and recognition of users among different social APPs are realized. According to the method, different social APP user portraits in the internet environment can be automatically analyzed and recognized, and different social APP accounts are associated to the same user according to the similarity degree of the user portraits. Compared with a traditional user discovery method, the method has the advantages that the accuracy of user recognition in different APPs is improved, and a relatively accurate entity relationship extraction method based on multi-target fusion is realized.

Description

technical field [0001] The invention relates to an entity relationship extraction method based on multi-object fusion, in particular to an entity relationship extraction method based on multi-object fusion based on a deep learning feature extraction model, and belongs to the technical field of natural language processing. Background technique [0002] In the 5G era, the emergence of social apps has changed people's lives. The social APP system has become the most popular mode of making friends. [0003] Social APPs provide users with the function of making friends with strangers, and users contain a lot of information in communication with strangers. Although social apps can meet user needs, most users use multiple social apps at the same time or have multiple accounts for the same social app. Taking China News as an example, the account name on Sina Weibo is CCTV China News, and the account name on WeChat Official Account is CCTV News. For authoritative organizations, bec...

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to aim at the discrete distribution of Internet public opinion and other related Internet speech records in the Internet. When using the existing Internet public opinion control mechanism to identify, there will be defects such as missing personal descriptions and low accuracy. A new method of entity relationship extraction based on multi-objective fusion is proposed to deal with technical problems such as different user cognition caused by different message gaps in APP, and low accuracy of user portrait discovery and matching.

Images