Information identification method based on weak supervised learning and related equipment

Abstract

The invention relates to the technical field of artificial intelligence, and provides an information identification method based on weak supervised learning. The method comprises the following steps:acquiring a plurality of common question answering FAQ data in a government affair scene, and constructing a first label data set based on the plurality of FAQ data; performing multi-label text classification on the first label data set by using a first convolutional neural network CNN to obtain a multi-label CNN model; acquiring a plurality of second label data sets, and adjusting the multi-labelCNN model by using the plurality of second label data sets to obtain a final multi-label classifier; receiving an input user question; inputting the user question into the final multi-label classifier to obtain a plurality of government affair entities matched with the user question; and outputting the plurality of government affair entities. The invention also relates to a blockchain technology,and the plurality of government affair entities can be uploaded to a blockchain. The method can be applied to the fields of smart government affairs, smart communities and the like needing information identification, so that the development of smart cities is promoted.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to an information identification method and related equipment based on weakly supervised learning. Background technique [0002] At present, in government affairs scenarios, various consultation questions are often encountered, and answers need to be provided. Government question answering robots play a key role in question answering. The government question-and-answer robot can assist government personnel in answering users' consultation questions, which usually correspond to relevant government entities, such as: registration of individual industrial and commercial household information, application for residence permits, approval of pesticide advertisements, funding for intellectual property rights protection, and issuance of breeding poultry production and operation licenses Wait. [0003] However, in practice, it is found that government entities are generally ...

AI Technical Summary

Problems solved by technology

[0003] However, in practice, it is found that government entities are generally more specialized, and different users have different colloquial names; in addition, the information of government entities is implicit in the consultation questions expressed by users, and it is difficult to accurately extract government entities. Make the user experience poor

Images