Method for calculating similarity of short texts by using deep convolution neural network
A similarity calculation and neural network technology, applied in the field of similarity calculation between short texts, can solve the problems of incomplete basis, keywords and concepts that cannot replace all texts, and cannot fully represent the similarity between two paragraphs of text, etc., to achieve simple calculation , the effect of improving the accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0034] The present invention will be further explained below in conjunction with specific embodiments and accompanying drawings.
[0035] The present invention comprises the following steps:
[0036] (1) Express several short texts into several matrices: first, select the words that appear in all relevant pages of knowledge fields on Wikipedia as the vocabulary; then, use the open source code of word2vec released by Google on the Internet to train the vocabulary, each Each word is expressed as a vector; finally, each word in the text is replaced in turn by the corresponding word vector in the vocabulary, each word vector occupies a row, and an ordered sequence of vectors is obtained, which can be regarded as a matrix, The number of lines is the number of words;
[0037] (2) Combine several short texts in pairs, and generate a similarity matrix for the matrix of two short texts in each group: first, for two paragraphs of text, take the two matrices corresponding to them in ste...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com