P
Prem Melville
Researcher at IBM
Publications - 54
Citations - 5004
Prem Melville is an academic researcher from IBM. The author has contributed to research in topics: Boosting (machine learning) & Feature (computer vision). The author has an hindex of 28, co-authored 54 publications receiving 4721 citations. Previous affiliations of Prem Melville include University of Texas at Austin.
Papers
More filters
Proceedings ArticleDOI
Content-boosted collaborative filtering for improved recommendations
TL;DR: This paper presents an elegant and effective framework for combining content and collaboration, which uses a content-based predictor to enhance existing user data, and then provides personalized suggestions through collaborative filtering.
Proceedings ArticleDOI
Sentiment analysis of blogs by combining lexical knowledge with text classification
TL;DR: This paper presents a unified framework in which one can use background lexical information in terms of word-class associations, and refine this information for specific domains using any available training examples, and shows that this approach performs better than using background knowledge or training data in isolation.
Proceedings ArticleDOI
Diverse ensembles for active learning
Prem Melville,Raymond J. Mooney +1 more
TL;DR: This paper introduces ACTIVE-DECORATE, which uses DECORATE committees to select good training examples and outperforms both Query by Bagging and Query by Boosting.
Proceedings ArticleDOI
Data Quality from Crowdsourcing: A Study of Annotation Selection Criteria
TL;DR: An empirical study is conducted to examine the effect of noisy annotations on the performance of sentiment classification models, and evaluate the utility of annotation selection on classification accuracy and efficiency.
Journal ArticleDOI
Creating diversity in ensembles using artificial data
Prem Melville,Raymond J. Mooney +1 more
TL;DR: A new method for generating ensembles, D ecorate (Diverse Ensemble Creation by Oppositional Relabeling of Artificial Training Examples), that directly constructs diverse hypotheses using additional artificially constructed training examples is presented.