A knowledge base question answering method and system in the field of electric power

Abstract

The invention provides a knowledge base question and answer method in the electric power field, including: performing part-of-speech tagging and syntactic analysis on the questions obtained in the electric power field to obtain the question sentence representation of the electric power field question; based on a pre-built key phrase recognition model Identify the question expression of the question in the electric power field, and obtain the key entities and key attribute phrases of the question in the electric power field; based on the key entities and key attribute phrases, search in the knowledge base of the electric power field to obtain the question. The key phrase recognition model includes: training the open domain question to obtain key entities and key attribute phrases corresponding to the open domain question, and the present invention avoids the high dependence on terms in the power domain model training in the prior art , the knowledge base question answering in the electric power field can be realized only by using the readily available open-domain resources.

Description

technical field [0001] The invention belongs to the fields of natural language processing and intelligent question answering, and in particular relates to a knowledge base question answering method and system in the electric power field. Background technique [0002] Domain-oriented knowledge base question answering technology is an important means to utilize knowledge information in this domain. Unlike traditional search engines that receive keywords as indexes, domain-specific knowledge base question answering technology receives natural language questions as the input of the system or device, and directly gives accurate and professional answers through domain knowledge base retrieval. The domain knowledge base question answering system or device can usually be used as an office assistant for a specific field, providing users with professional domain knowledge retrieval services, and has been widely used in work and life, such as: Microsoft's "Cortana", Alibaba's " Cube S...

AI Technical Summary

Problems solved by technology

However, handwriting templates is a process with high labor and time costs, especially for complex knowledge bases, it is difficult to cover all question forms or knowledge base content, and it is incomplete; the end-to-end method does not have a handwriting template process, The cost is low, but it usually requires a large amount of Q&A data related to the content of the knowledge base to train the model, which has strong domain dependence, but the Q&A data is difficult to obtain or organize

Images