Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
FEATURE EXTRACTION METHOD AND APPARATUS FOR TEXT CLASSIFICATION
Document Type and Number:
WIPO Patent Application WO/2022/166830
Kind Code:
A1
Abstract:
Provided are a feature extraction method and apparatus for text classification. The method comprises: using multiple feature extraction methods on text data to extract features so as to obtain feature items (101); determining the classification accuracy of the multiple feature extraction methods under different feature item dimension values on the basis of the feature items corresponding to the multiple feature extraction methods, the feature item dimension values being used to represent the number of feature items (102); determining the optimal classification accuracy and optimal feature dimension value corresponding to each feature extraction method on the basis of the classification accuracy of the multiple feature extraction methods (103); determining a target feature extraction method and a target feature dimension value corresponding to the text data on the basis of the optimal classification accuracy and optimal feature dimension value corresponding to each feature extraction method (104). The use of the provided method helps to increase the accuracy of text classification.

Inventors:
HUO XIAOQIAN (CN)
Application Number:
PCT/CN2022/074714
Publication Date:
August 11, 2022
Filing Date:
January 28, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
BEIJING UNISOC COMM TECH CO LTD (CN)
International Classes:
G06F16/35
Foreign References:
CN112989036A2021-06-18
CN110597878A2019-12-20
CN111666748A2020-09-15
CN107766792A2018-03-06
US20200394557A12020-12-17
US20190377823A12019-12-12
Attorney, Agent or Firm:
SCIHEAD IP LAW FIRM (CN)
Download PDF: