Leveraging Unlabeled Data to Scale Blocking for Record Linkage.
Yunbo CaoZhiyuan ChenJiamin ZhuPei YueChin-Yew LinYong YuPublished in: IJCAI (2011)
Keyphrases
- record linkage
- unlabeled data
- labeled data
- semi supervised learning
- semi supervised
- co training
- active learning
- data cleaning
- duplicate detection
- privacy preserving
- text classification
- semi supervised classification
- supervised learning
- data points
- labeled examples
- labeled and unlabeled data
- class labels
- linked data
- learning algorithm
- training set
- text categorization
- number of labeled examples
- labeled training data
- training data
- small set of labeled
- training examples
- positive examples
- supervised and semi supervised
- pairwise
- prior knowledge
- label propagation
- machine learning
- transfer learning
- supervised learning algorithms
- domain adaptation
- unsupervised learning
- text mining
- unlabeled instances
- decision boundary
- multi view
- labeled data for training