A supervised scheme for aspect extraction in sentiment analysis using the hybrid feature set of word dependency relations and lemmas.
Bhavana R. BhamareJeyanthi PrabhuPublished in: PeerJ Comput. Sci. (2021)
Keyphrases
- sentiment analysis
- feature set
- dependency relations
- sentiment words
- feature selection
- sentence level
- text classification
- opinion mining
- sentiment lexicon
- sentiment classification
- classification accuracy
- feature vectors
- feature space
- word sense disambiguation
- semantic relations
- positive or negative
- feature extraction
- feature selections
- text mining
- text categorization
- semantic features
- natural language processing
- supervised learning
- n gram
- semi supervised
- information extraction
- co occurrence
- product reviews
- online reviews
- machine learning
- feature subset
- product features
- learning algorithm
- sentiment polarity
- neural network
- databases
- opinion words
- domain knowledge