Improving the results of string kernels in sentiment analysis and Arabic dialect identification by adapting them to your test set.
Radu Tudor IonescuAndrei M. ButnaruPublished in: CoRR (2018)
Keyphrases
- test set
- sentiment analysis
- string kernels
- opinion mining
- error rate
- text classification
- sentence level
- sentiment classification
- training set
- computational biology
- grammatical inference
- natural language processing
- spam filtering
- training data
- product reviews
- feature selection
- feature extraction
- natural language
- data sets
- text mining
- object detection
- support vector machine