Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Chinese intelligent question and answer system method based on word similarity of a network platform

An intelligent question answering and network platform technology, applied in text database query, unstructured text data retrieval, special data processing applications, etc., can solve problems such as poor answer extraction effect, inaccurate similarity, and only considering statistical frequency, etc.

Inactive Publication Date: 2019-04-05
ZHEJIANG NORMAL UNIVERSITY
View PDF2 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0018] The disadvantage of this method is that the semantic analysis method relies on supervised data such as for vocabulary tagging and model training. Due to the requirements of data annotation, it can usually only be used in specific fields and requires manual labor. Label a large number of logical expressions for training
[0022] The disadvantage of this existing method is that when calculating classification or calculating similarity, most of them use large-scale corpus for statistical calculation.
[0026] The disadvantage of this method is that this method only considers the statistical frequency of words, but ignores the linguistic meaning of words. Obviously, the similarity calculated based on this method is not accurate; the results obtained according to the statistical method are greatly disturbed by data sparseness. , so there will be obvious calculation errors
[0033] Difficulty: The existing technology needs to label a large amount of manual training data, the situation where the similarity calculation of short texts in the question answering system is not suitable for statistical methods, and the situation where the language structure of the question sentence and the answer is far apart when the pattern matching is far away, and the answer extraction effect is poor

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Chinese intelligent question and answer system method based on word similarity of a network platform
  • Chinese intelligent question and answer system method based on word similarity of a network platform
  • Chinese intelligent question and answer system method based on word similarity of a network platform

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0158] 1. In order to verify the effect of the word similarity calculation method, the present invention carries out the experiment of similarity calculation with several groups of commonly used words. In the experiment, two word similarity calculation methods are used for verification, which are the method proposed by the present invention and an existing method.

[0159] The present invention assumes that the two methods have different algorithms when calculating the sememe set, that is, the method proposed by the present invention is the same as the existing method that only takes the maximum value, and other steps are the same. Table 4 shows the similarity calculation results of some words.

[0160]

[0161]

[0162] 2. Performance test:

[0163] In the experiment, 6000 nouns were tested, and the consistency rate was compared between the automatically recognized semantic class results and the manually proofreaded semantic classes in the CSD dictionary. The results a...

Embodiment 2

[0167] 1. In order to verify the application of word similarity in the question answering system, the present invention has established a knowledge base answering system with nearly 3000 pieces of data, involving various fields. Examples include books, movies, people and places of interest.

[0168] For all questions, and all answers in the knowledge base question answering system, word division is realized based on the software ICTCLAS (http: / / www.ICtcas.org / ). After that, stop words and symbols are removed.

[0169] Specific names of people, places, and time that do not appear in HowNet are replaced with abstract words such as "person's name", "place name", and "time" respectively. For example, replace "Yang Feiyu" with "person's name"; replace "Mugecuo" with "place name"; replace "2018 / 1 / 30" with "time".

[0170] Remove repetitions.

[0171] Table 6 shows some example questions and answers

[0172] Example Questions and Answers for Part 6 of Table

[0173]

[0174] ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the technical field of network natural language processing. The invention discloses a Chinese intelligent question and answer system method based on word similarity of a network platform. In a knowledge base question and answer system, each question and each answer are regarded as two word sets, each word in the question set is matched with each word in the answer set, theword similarity is calculated, then the maximum similarity value is obtained, and then the average value of the maximum value is obtained; The method is simple and high in efficiency; According to the method, the problem of data sparsity of an existing vector included angle cosine method is solved; Meanwhile, the situation of inaccurate answer extraction caused by inconsistent question and answerlanguage structures in the existing mode matching method is also overcome; According to the word similarity algorithm, answers can be found in the knowledge base question and answer system more reasonably and efficiently.

Description

technical field [0001] The invention belongs to the field of intelligent question answering systems for natural language processing, in particular to a Chinese intelligent question answering system method based on word similarity of a network platform. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: [0003] The calculation of word similarity is to use a specific value to represent the similarity between two words. It is the main means of understanding the semantic information of words and one of the basic tasks of natural language processing. Word similarity calculation is the main method for semantic understanding, and the solution to the similarity problem will promote the development of related application technologies in the field of natural language processing, such as information retrieval, word sense disambiguation, machine translation, and question answering systems. [0004] Among them, the intellig...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F16/332G06F16/33
Inventor 聂红梅虞协俊周家庆
Owner ZHEJIANG NORMAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products