Home › Categories › Technology› Twitter Sentiment Transfer

Twitter Sentiment Transfer.pdf

JT-DH_2020_Robnik-Sikonja-et-al_Cross-lingual-Transfer-of-Twitter-Sentiment-Models.pdf

🔗 Source: nl.ijs.si

📊 Size: 742 KB

📄 Pages: 6 pages

⬇️ Downloads: 135

Summary

Cross-lingual word embeddings map words from different languages to a common vector space, enabling transfer of machine learning models between languages. The authors use cross-lingual word embeddings to transfer Twitter sentiment prediction models between 13 languages, focusing on two transfer mechanisms: transferring trained models and expanding training sets with instances from other languages. Experiments show that transferring trained models between similar languages is effective, while expanding training sets did not increase predictive performance. The authors also compare the common vector space constructed by the LASER library with multilingual BERT and traditional bag-of-words approaches, finding a relatively low decrease in predictive performance when transferring trained sentiment prediction models between languages.

Download File

Description

Cross-lingual word embeddings map words from different languages to a common vector space, enabling transfer of machine learning models between languages.

Technical Information

File Format: PDF
File Size: 742 KB
Pages: 6
Language: EN
Total Downloads: 135
Last Updated: 3 hours ago

Document Overview

This PDF document about Twitter Sentiment Transfer provides comprehensive information and guidance. Whether you're a beginner or advanced user, this resource offers valuable insights into Twitter Sentiment Transfer.