Projects

• Speech
  Speaker identification on a large datasets
  Voice conversion
  Speech summarization
  
  
• Linguistic Networks
  Unsupervised discovery of function words in a language
  Computational Study of Bollywood song lyrics
  
  
• Parsing
  Malt, MST and Bidirectional parsers
  Iterative learning for POS tagging, chunking, NER and parsing
  
  
• Semantics
  Sanity checks for PurposeNet
  Populating Action Ontology using Verbnet
  Mapping semantic roles to karaka relation using Verbnet
  
  
• Morphology
  Building a morphological analyzer
  
  
• Anaphora Resolution
  Anaphora resolution for Hindi
  
  
• Question Generation
  Question Generation from paragraph
  
  
• Resource Creation
  Creating linguistic resource of different granularities
  
  
• Transfer Based MT (TBMT)
  Transfer based MT between Indian languages
  
  
• Statistical MT
  Basic experiments in SMT
  
  
• Sentiment Analysis
  Development of Emotion Lexicon and Word Level Emotion Analysis System
  Development of SentiWordNet and Sentiment Classification
  
  
• Information Extraction, Information Retrieval (IE/IR)
  Amazon data (search customer reviews)
  Twitter data clustering based on categories
  
  Details