A method of problem equivalence discrimination based on semi-supervised learning combined with ensemble learning

Abstract

The invention discloses a method for judging the equivalence of a problem by combining semi-supervised learning with integrated learning, including S1, normalization of synonyms: 1) word vector embedding; 2) word similarity judgment; 3) manual judgment; S2, semantics Equivalence recognition: calculation of dual network based on LSTM; calculation of dual network model based on CNN; calculation based on Match Pyramid model; manual feature extraction. The present invention generates and introduces synonyms in a semi-supervised manner, so that the entire system can be flexibly adjusted according to specific fields, and four different models are used to learn the semantic equivalence of the problem to judge the semantic equivalence of the problem, thereby giving full play to the advantages of different models.

Description

Technical field[0001]The present invention relates to the field of data mining techniques, and more particularly to a half-supervised learning to combine malicious discriminant in the problem of integrated learning.Background technique[0002]The intelligent question and answer system uses a question-and-answer form, the accurate positioning user needs, and provides users with personalized information services. With the development of artificial intelligence technology, the intelligent question and answer system has also gained more and more applications in banks, insurance, services, and government.[0003]Intelligent Q & A system can generally be divided into two categories for Domain Specific or Generaldomain. For the former, due to the accurate answer must be based on the expertise in this field, the intelligent question and answer system generally depends on "Question - Answer" Knowledge Base. Therefore, if equivalent issues are returned by equivalent problems through effective ide...

AI Technical Summary

Problems solved by technology

On the one hand, it is difficult for this method to describe the equivalence / inequivalence in the deep semantics of the problem

On the other hand, it takes a lot of time and effort to manually construct a dictionary of synonyms, and it is difficult to exhaust the possibilities of synonyms

In addition to the traditional problem equivalence judgment method, although the method based on deep learning avoids the above problems in a sense, it is difficult to incorporate business prior knowledge, so it is difficult to adjust according to specific fields

Images