Categorized Projects

1] Summarization/Information retrieval -

• information extraction
• blog retrieval
• sentiment extraction
• information extraction for biology
• news summarization
• passage retrieval
• summarization of papers
• searching compressed text
• social search engines
• summarizing scientific papers
• linguistically motivated shingling
• summarizing news in a foreign language
• combining LSI/SVD with dependency trees
• building document closures for indexing
• applying LSI/SVD for summarization
• clustering short sentences by similarity
• implementing shingling for document duplicate and near-duplicate recognition
• clustering scientific papers by topic using citation information
• clustering search results
• summarizing news
• evolving summarization
• graph based summarization
• identifying paraphrases of facts
• extracting generifs
• aligning summaries w/stories w/headlines
• information extraction using mallet (CRF)
• Compress the text of Wikipedia (losslessly).
• event extraction
• extracting bilingual vocabularies
• semi-supervised IR
• tree kernels for IR
• extrtating social networks from text
• AAN paper retrieval
• collect URLs from papers
• cross language IR and SVD
• extracting protein interactions
• indexing over evolving text
• web IR task of TREC
• blog IR task of TREC
• paper retrieval using the ACL Anthology

<2] Query modification/QA -

• question answering
• automatic query correction/expansion
• query completion/recommendation
• extract names of people and their descriptions from the web
• Spelling correction using query logs.
• Automatic query expansion.
• automatic query correction/expansion

3] Analysis/Learning/Identification -

• web graph analysis
• language identification
• finding similar documents
• answer validation (clef)
• spam recognition
• relation recognition
• blog analysis
• query log analysis
• person identification
• named entity recognition
• graph-based semi-supervised learning on small world networks
• personal name disambiguation
• semisupervised learning
• lexical homophily
• tree kernels
• semantic kernels
• combining SVD and kernels (e.g., for syntax)
• factuality recognition
• learning on small worlds
• identifying high impact papers
• identifying communities in paper networks
• recognizing anachronistic sentences
• movie recommendations
• predicting what page the user will click on next
• identifying redundant information
• dependency kernels
• recommendation system for documents
• analysis of congressional data
• network analysis of people and places
• tumbl semi-supervised learning on graphs
• analysing the enron corpus
• network analysis of some corpus
• modeling entity burstiness
• extract names of people and their descriptions from the web

4] Misc -

• paraphrasing and entailment
• clairlib extensions
• document models
• document closures
• word distribution across hyperlinks
• textual entailment recognition
• modeling entity burstiness
• information diffusion on blogs (e.g., who are the trendsetters)
• URL recommendation site
• identifying trends in blogs
• analyze large datasets
• adversarial IR (spam)
• stability of pagerank
• correlate pagerank with other metrics
• sentiment analysis (movies or people, etc)
• community finding
• information diffusion on blogs
• grammatically accurate edit distance
• modeling burstiness
• language modeling for IR
• domain-topic models
• NERC from Web
• table recognition and parsing
• measuring the size of columbia.edu
• web based spelling correction
• selling ads
• lexical centrality
• compare burstiness models
• INEX wikipedia
• INEX (IEEE papers)
• learning translations
• approximate record matching
• sentence decomposition
• comparison of Lucene, Indri, Lemur, etc.
• web based language model smoothing
• porting clairlib to another language
• other additions to clairlib
• look at some distribution like Benford
• incorporate syntax into SVD
• lexical networks
• verify bow-tie model
• self triggering across hyperlinks
• dynamic lexrank
• community finding on graphs