Text Emotion Classification Research Based on Improved Latent Semantic Analysis Algorithm

DOI

10.2991/iccsee.2013.55How to use a DOI?

Keywords

Latent Semantic Analysis, Vector Space Model, Text Emotion Classification,

Abstract

The emotion classification of text is an important research direction of text mining. Application on emotion text classification, latent semantic analysis algorithm has advantage of small occupied space, applicable to a large scale of text classifications. Compared with the traditional vector space model, latent semantic analysis algorithms reduce the search space for text classification by means of singular value decomposition for term and document matrix. Moreover, latent semantic analysis algorithms solve the problem of words with multiple meanings by analyzing the term at the semantic level. Using an improved latent semantic analysis algorithm to classify the test set by their emotion. The new cluster centroid is the average vector for each emotion category, and access to emotions classification for training dataset by calculating similarity of the average vector and test textual. The experimental results show that the improved latent semantic analysis algorithm have high precision and recall rate as same as the original algorithm, the efficiency of text emotion classification improved 4 percentage points.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xuren Wang
AU  - Qiuhui Zheng
PY  - 2013/03
DA  - 2013/03
TI  - Text Emotion Classification Research Based on Improved Latent Semantic Analysis Algorithm
BT  - Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013)
PB  - Atlantis Press
SP  - 210
EP  - 213
SN  - 1951-6851
UR  - https://doi.org/10.2991/iccsee.2013.55
DO  - 10.2991/iccsee.2013.55
ID  - Wang2013/03
ER  -