Sentiment Analysis and Topic Classification based on Binary Maximum Entropy Classifiers.
Fernando BatistaRicardo RibeiroPublished in: Proces. del Leng. Natural (2013)
Keyphrases
- maximum entropy
- sentiment analysis
- class conditional
- text classification
- supervised classifiers
- kernel logistic regression
- collective classification
- binary classifiers
- maximum entropy principle
- decision trees
- classification algorithm
- opinion mining
- supervised classification
- support vector
- sentiment classification
- class labels
- named entity recognition
- public opinion
- product reviews
- feature selection
- training set
- machine learning algorithms
- training samples
- user generated
- machine learning
- naive bayes
- text mining
- conditional random fields
- decision boundary
- training data
- feature space
- social media
- sentence level
- class probabilities
- multi class
- supervised learning
- natural language processing
- image classification
- sentiment lexicon
- sentiment polarity
- hidden markov models
- multi document summarization
- topic modeling