Micro-blog emotion prediction method based on weak supervised type multi-modal deep learning
A technology of deep learning and prediction methods, applied in prediction, biological neural network model, other database retrieval and other directions, can solve the problems of multi-modal discriminant representation and limited data labels, and achieve the effect of performance improvement and effect improvement
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0023] The following embodiments will further illustrate the present invention in conjunction with the accompanying drawings.
[0024] Embodiments of the present invention include the following steps:
[0025] Step 1 Microblog multimodal data preprocessing, the specific method is as follows:
[0026]First, de-duplicate the crawled Weibo data, then filter out the label symbols and external links in the Weibo content, and finally use the Chinese Academy of Sciences automatic word segmentation tool ICTCLAS to segment the Weibo text content (Text segment); As a noise label (emoji category), first collect all the emoticons in the text corpus, then screen out 49 frequently used emoticons, and finally build an emoji word bag model for each Weibo as an emoji category (Bag- of-emoticon-word) to obtain clean labels by manually labeling the emotional polarity of microblog data (emotional polarity categories: positive, negative, and neutral).
[0027] Step 2 Weakly supervised training o...
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