Paper Info
Title
An Empirical Study on Machine Learning-Based Sentiment Classification Using Polarity Clues.
Abstract
In recent years a variety of approaches in classifying the sentiment polarity of texts have been proposed. While in the majority of approaches the determination of subjectivity or polarity-related term features is at the center, the number of publicly available dictionaries is rather limited. In this paper, we investigate the performance of combining lexical resources with machine learning-based classifier for the task of sentiment classification. We systematically analyze four different English and three different German polarity dictionaries as a resources for a sentiment-based feature selection. The evaluation results show that smaller but more controlled dictionaries used for feature selection perform within a SVM-based classification setup equally good compared to the biggest available resources.
Year
DOI
Venue
2010
10.1007/978-3-642-22810-0_15
Lecture Notes in Business Information Processing
Keywords
Field
DocType
Machine learning,Support vector machine,Sentiment analysis,Polarity classification,Polarity resources
Data mining,Feature selection,Subjectivity,Sentiment analysis,Computer science,Support vector machine,Artificial intelligence,Classifier (linguistics),Machine learning,Empirical research,German
Conference
Volume
ISSN
Citations 
75
1865-1348
2
PageRank 
References 
Authors
0.36
29
1
Name
Order
Citations
PageRank
Ulli Waltinger16410.76