Triplex transfer learning: exploiting both shared and distinct concepts for text classification.
Fuzhen ZhuangPing LuoChangying DuQing HeZhongzhi ShiPublished in: WSDM (2013)
Keyphrases
- transfer learning
- text classification
- labeled data
- learning tasks
- text categorization
- knowledge transfer
- machine learning
- text mining
- cross domain text classification
- sentiment analysis
- feature selection
- target domain
- unsupervised learning
- multi task
- naive bayes
- text classifiers
- manifold alignment
- cross domain
- transfer knowledge
- unlabeled data
- sentiment classification
- co training
- domain adaptation
- document classification
- multi task learning
- knn
- multi label
- machine learning algorithms
- semi supervised learning
- k nearest neighbor
- active learning
- nearest neighbor
- previously learned
- data sets
- neural network
- real world
- learning algorithm
- decision trees
- reinforcement learning
- collaborative filtering
- supervised learning