Enhanced Feature Selection Using Word Embeddings for Self-Admitted Technical Debt Identification.
Jernej FlisarVili PodgorelecPublished in: SEAA (2018)
Keyphrases
- feature selection
- dimensionality reduction
- text categorization
- feature selection algorithms
- document frequency
- text classification
- n gram
- pointwise mutual information
- term frequency
- feature subset
- short term
- feature set
- co occurrence
- multi class
- feature extraction
- feature space
- machine learning
- vector space
- high dimensionality
- classification accuracy
- manifold learning
- information gain
- feature weighting
- risk management
- word segmentation
- discriminative features
- keywords
- multi task
- mutual information
- model selection