Parsimonious unsupervised and semi-supervised domain adaptation with good similarity functions.
Emilie MorvantAmaury HabrardStéphane AyachePublished in: Knowl. Inf. Syst. (2012)
Keyphrases
- domain adaptation
- similarity function
- semi supervised
- distance function
- pairwise
- similarity measure
- multiple sources
- cross domain
- semi supervised learning
- labeled data
- transfer learning
- sentiment classification
- kernel methods
- unlabeled data
- maximum margin
- test data
- reproducing kernel hilbert space
- prior knowledge
- text classification
- knn
- target domain
- training data
- unsupervised learning
- k nearest neighbor
- document classification
- co training
- feature vectors
- feature selection
- data mining