A Chinese Question Answering System Based on Neural Network

A neural network and question answering system technology, applied in the field of information retrieval, can solve problems such as narrow coverage, low system accuracy and recall rate, user experience damage, etc., to avoid space overhead and improve accuracy.

CN101566998BInactive Publication Date: 2011-12-28HUAZHONG NORMAL UNIV
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2011-12-28
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a Chinese question answering system based on a neural network, which includes a user interface module, a question pre-segmentation module, a neuron pre-marking module, a learning and training module, a neuron knowledge base module, a semantic block recognition module, a question set index module and Answer reasoning module. Through this system, the SIE encoding method is first used to encode the words in the semantic block according to their relative positions; then the problem of identifying the semantic block of the question sentence is transformed into a label classification problem, and the classification model based on the neural network is used to determine the semantic structure of the question sentence ;Finally, combined with the semantic structure of the question, realize the similarity calculation of the question based on the neural network, and compare the weights between the semantic features of the question by extracting the semantic features of the marked question to provide a basis for the final answer reasoning. The invention integrates the syntax, semantics and context knowledge of sentences, and can simulate the process of processing sentences by humans.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of information retrieval, in particular to a Chinese question answering system based on a neural network. Background technique

[0002] Question Answering System (Question Answering System) refers to a system that can answer questions that are input by computer users and described in natural language, and is generally implemented by computer software system technology. The question answering system integrates natural language processing, information retrieval, and knowledge representation, and is increasingly becoming a research hotspot in the world. It not only allows users to ask questions in natural language, but also returns users a concise and accurate answer instead of some related web pages. Therefore, compared with traditional search engines that rely on keyword matching, the question answering system can better meet the user's retrieval needs and find out the answers that users need more accurately...

Examples

Embodiment Construction

[0029] Below in conjunction with accompanying drawing and example the present invention is described in further detail.

[0030] Such as figure 1 As shown, the Chinese question answering system provided by the embodiment of the present invention includes a user interface module 10, a question pre-segmentation module 20, a neuron pre-marking module 30, a learning and training module 40, a neuron knowledge base module 50, a semantic block recognition module 60, The question set indexing module 70, the answer reasoning module 80 and the additional sample supplementary learning module 90, each module can be implemented by those skilled in the art using computer software technology according to the technical solution of the present invention.

[0031] The user interface module 10 is used to realize user interaction, including two functions: one is to accept user input questions, then check the expression of the user input questions, and submit the checked user input questions to th...