Abstract
The aim of this paper is to evaluate the use of content and style features in automatic classification of intentions of Tweets. For this we propose different style features and evaluate them using a machine learning approach. We found that although the style features by themselves are useful for the identification of the intentions of tweets, it is better to combine such features with the content ones. We present a set of experiments, where we achieved a 9.46 % of improvement on the overall performance of the classification with the combination of content and style features as compared with the content features.
Original language | English |
---|---|
Pages (from-to) | 120-128 |
Number of pages | 9 |
Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Volume | 8864 |
DOIs | |
State | Published - 2014 |
Keywords
- Detection of intention
- Short texts
- Text classification