Semi-supervised text sentiment classification method based on random feature subspace
A random subspace and random feature technology, applied in the field of semi-supervised text sentiment classification based on random feature subspace, can solve the problems of small classifier difference and large misclassified samples.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0056] The present invention preprocesses comment texts to construct a global feature set, and expresses all comment texts into vector form, and then marks the emotional polarity of some comment texts to obtain a marked sample set and an unmarked sample set; and then uses Lasso The method calculates the feature weights of all feature words in the global feature set, and uses the feature weights as probability to extract some feature words to construct a random subspace, maps the marked sample set to the random subspace and trains the classifier, and uses the unlabeled sample set to perform Collaborative training to get the final classifier; finally integrate Z classifiers in the form of main voting, and obtain the final integrated classifier F(x ε ). Specifically, as figure 1 Shown, the inventive method comprises the following steps:
[0057] Step 1. Construct a global feature set T:
[0058] Step 1.1. Obtain n comment texts to form a comment text set D, denoted as D={d 1 ...
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