Training method and device of question-answer pair classification model

Abstract

The embodiment of the invention provides a training method and device of a question-answer pair classification model. The method comprises the steps of obtaining question-answer pair data; extracting question-answer pair characteristics from the question-answer pair data; labeling a classification tag on the question-answer pair data according to the quality of the question-answer pair data; training the question-answer pair classification model through the question-answer pair characteristics and the classification tag. The quality of the question-answer pair data is adopted to automatically label a large quantity of training sets, the question-answer pair classification model is classified, that is, to predict the quality score, the artificial strategy is avoided, thus the problems that feature information adopted in the artificial strategy is few, the active feedback rate of a user is low, the subjective judgment of a quizzer is depended on, the cheating phenomenon in the advertisement is serious, and the strategy is unstable caused when the feedback information of the user to new question-answer pair data and historical question-answer pair data is imbalanced are solved, and good forecast accuracy rates are obtained in both the historical question-answer pair data and the new question-answer pair data.

Description

technical field [0001] The invention relates to the technical field of computer processing, in particular to a question-answer pair classification model training method and a question-answer pair classification model training device. Background technique [0002] At present, there are many interactive question-and-answer platforms on the Internet. Users ask their own questions on the question-and-answer platform, and the question-and-answer platform mobilizes other users to answer and solve the questions of the questioners. [0003] The question-and-answer platform has accumulated a large number of users and generated a large amount of question-and-answer data (that is, questions and answers). Among them, the quality of the question-and-answer data varies. High-quality question-and-answer data is an important data resource for the question-and-answer platform. [0004] In order to dig out high-quality question-answer pair data, the traditional method is to calculate the qua...

AI Technical Summary

Problems solved by technology

[0006] However, the artificial strategy uses less feature information, the user’s active feedback rate is low, and it relies on the subjective judgment of the questioner. The phenomenon of advertising cheating is serious. The user’s feedback information on the new question-answer data and historical question-answer data is unbalanced, resulting in an unstable strategy. , leading to a lower accuracy rate of question answering on the data

[0007] In particular, the newly generated question-answer pair data, because of the lack of user feedback, the accuracy rate of the question-answer pair data is lower

Images