Transferring knowledge with source selection to learn IR functions on unlabeled collections.
Parantapa GoswamiMassih-Reza AminiÉric GaussierPublished in: CIKM (2013)
Keyphrases
- source selection
- distributed information retrieval
- transferring knowledge
- labeled data
- transfer learning
- information retrieval
- knowledge transfer
- resource selection
- multitask learning
- multiple tasks
- unlabeled data
- active learning
- semi supervised
- semi supervised learning
- data fusion
- information filtering
- training data
- document collections
- support vector machine
- federated search
- prior knowledge
- unsupervised learning
- text classification
- information retrieval systems
- supervised learning
- relevance feedback
- text retrieval
- high dimensional
- effective retrieval
- digital libraries
- machine learning