Systems and methods for performing contextual classification using supervised and unsupervised training

作者: Vineet Mahajan , Thu Kyaw , Sang Chui Song , Elena Haliczer

DOI:

关键词:

摘要: Computerized systems and methods are disclosed for performing contextual classification of objects using supervised unsupervised training. In accordance with one implementation, content reviewers may review training submit data preprocessing analysis. The be preprocessed to identify key terms phrases, such as by stemming, tokenization, or n-gram analysis, form vectorized objects. used train more models subsequent certain implementations, training, among other steps, performed in parallel over multiple machines improve efficiency. a wide variety applications, article moderation.

参考文章(19)
D. Sculley, On Free Speech and Civil Discourse: Filtering Abuse in Blog Comments. conference on email and anti-spam. ,(2008)
Patrick Haffner, Nicholas Duffield, Jeffrey Erman, Subhabrata Sen, Yu Jin, Scalable traffic classifier and classifier training system ,(2009)
Yunbo Cao, Jian-Tao Sun, Ming Zhou, Shen Huang, Ling Bao, Zheng Chen, Jian Wang, Christoph R. Ponath, Chin-Yew Lin, Smart Sentiment Classifier for Product Reviews ,(2007)
Marko Krema, Rayid Ghani, Sentiment classifiers based on feature extraction ,(2011)
Kenneth R. Wood, Samuel Gavin Smyth, Hilary Lyndsay Williams, Georgina E. Browne, Stephen E. Hodges, David Alexander Butler, Emma L. Berry, Chris Demetrios Karkanias, Memory training via visual journal ,(2007)
Jason Weston, Isabelle Guyon, Stephen Barnhill, Methods of identifying biological patterns using multiple data sets ,(2000)
Lili Diao, Jackie Cao, Vincent Chan, Lightweight content filtering system for mobile phones ,(2006)