We propose and test an objective criterion for evaluation of clustering performance: How well does a clustering algorithm run on unlabeled data aid a classification algorithm? The...
For manyknowledgeintensive applications, it is necessary to have extensive domain-specific knowledgein addition to general-purpose knowledge bases usually built around MachineRead...
This paper presents an approach to detection of the semantic types of relation arguments employing the WordNet hierarchy. Using the SemEval-2007 data, we show that the method allo...
In this paper, we present a new co-training strategy that makes use of unlabelled data. It trains two predictors in parallel, with each predictor labelling the unlabelled data for...
Terms used in search queries often have multiple meanings. Consequently, search results corresponding to different meanings may be retrieved, making identifying relevant results in...