Unlock instant, AI-driven research and patent intelligence for your innovation.
A method and device for semi-supervised field word mining and classification
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A domain word and semi-supervised technology, applied in text database clustering/classification, character and pattern recognition, text database query, etc., can solve problems such as poor effect and difficulty in obtaining labeled corpus
Active Publication Date: 2020-04-10
广东惠禾科技发展有限公司
View PDF3 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
However, supervision requires a large amount of labeled corpus, and labeled corpus is actually difficult to obtain, so the actual use effect is not good.
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
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0080] Embodiment 1 of the present invention discloses a method of semi-supervised field word mining and classification, such as figure 1 shown, including the following steps:
[0081] Step 101, perform word segmentation and syntactic analysis on the text data in the field to be processed, and obtain the word vector matrix of all words in the text data based on the result of the word segmentation;
[0082] Specifically, in the field of medicine, for example, text data can be obtained from medical websites through web crawlers, etc. Text data in other fields is similar, as long as the corresponding text data can be obtained, it is not limited to specific methods.
[0083] After obtaining the text data, word segmentation and syntactic analysis will be performed;
[0084] As for the "obtaining the word vector matrix of all words in the text data based on the result of the word segmentation" in the above steps includes:
[0085] Obtaining the result of word segmentation of the t...
Embodiment 2
[0115] Embodiment 2 of the present invention discloses a semi-supervised field word mining and classification equipment, such as figure 2 shown, including:
[0116] An acquisition module 201, configured to perform word segmentation and syntactic analysis on the text data in the field to be processed, and obtain word vector matrices of all words in the text data based on the result of the word segmentation;
[0117] The construction module 202 is used to start with a certain number of seed words artificially constructed in the text data, expand the seed words based on the part-of-speech and syntactic composition mode of the seed words in the text data, and use word frequency, part-of-speech , word vectors to filter the seed words to obtain the seed vocabulary;
[0118] Generating module 203, for described seed vocabulary, utilize word vector, knowledge base, statistical feature etc. to determine the general similarity of any two words, and generate word similarity matrix with...
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
Login to View More
Abstract
Embodiments of the invention disclose a semi-supervised field word mining and classifying method and equipment. The method comprises the following steps of: preprocessing a field-related corpus and constructing a seed word list and a word similarity matrix; mining candidate field words and determining similarity distribution of the candidate field words; and carrying out category classification onthe screened field words. According to the method and the equipment, a semi-supervised manner is adopted, so that the field word mining and classification can be completed on the basis of common field texts and a small amount of seed word tables without a large amount of tagged data.
Description
technical field [0001] The invention relates to the field of domain word mining and classification, in particular to a method and equipment for semi-supervised domain word mining and classification. Background technique [0002] Domain words are the characteristics that can best represent the characteristics of the domain and distinguish other domains, and domain words can be divided into different category labels according to different roles in the domain. Domain words and their categories constitute the basic vocabulary data of the domain; therefore, domain words The mining and classification of Chinese information processing is an important basic work in Chinese information processing. In many Chinese information processing projects (such as automatic question answering, automatic summarization, automatic classification, search engines, etc.), domain word mining and classification problems will be involved. [0003] At present, the mining and classification algorithms of ...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More