Transcript Natural Language Processing Group - AU

Natural Language Processing
Group
Computer Sc. & Engg. Department
JADAVPUR UNIVERSITY
KOLKATA – 700 032, INDIA.
Professor Sivaji Bandyopadhyay
[email protected]
Research Areas

Natural Access to Internet & Other Resources
– Headline Generation
– Headline Translation
– Document Translation
– Multilingual Multidocument Summarization

Cross-lingual Information Management
– Multilingual and Cross-lingual IR
– Open Domain Question Answering
Natural Access to Internet &
Other Resources

Headline Generation
– A machine translation problem
 the input document identified by a set of features
and output headline represents some of them
– Example Base
 Set of features in the input document and the
headline template(s)
– Implemented for generating headlines from
cricket news in English
Natural Access to Internet &
Other Resources

Headline Translation
– A Hybrid MT system for translating English news
headlines to Bengali
– Syntactic and Semantic classification of news headlines
done
– Anaphora and Coreference classes identified in news
headlines
– Translation Strategy
The input headline first searched in
Translation Memory, else tagged and searched in
Tagged Example Base, else analyzed and matched in
Phrasal Example Base, else heuristics applied
Natural Access to Internet &
Other Resources

Document Translation
– Prototype developed for A Hybrid MT system from
English to Bengali
– Translation Strategy




Identify the constituent phrases of a sentence using a Shallow
Parser
translate them individually using an Example Base
arrange the translated phrases using heuristics to form the
target language output
Verb phrases translated using Morphological Paradigm Suffix
Tables
Natural Access to Internet &
Other Resources

Multilingual Multidocument Summarization
– Multidocument summarization in each language


Summarize one of the documents using extraction methods
Revise the summary using other documents
– Summary in the target language is the reference
summary
– Translate all summaries to the target language
– Revise the reference summary
Cross-lingual Information
Management

Multilingual and Cross-lingual IR
– A Cross Language Database (CLDB) System in Bengali
–
–
–
–
and Hindi developed
Natural language query analyzed using a Template
Grammar and Knowledge Bases to produce the
corresponding SQL statement
Cooperative response in the query language
Anaphora / Coreference in CLDB studied
Database updates and elliptical queries also supported
Cross-lingual Information
Management

Open Domain Question Answering
– Work being done for English
– Currently building a set of question templates
(Qtargets) and the corresponding Answer patterns with
relative weights
– Input question analyzed to produce the corresponding
question template


Appropriate answer pattern retrieved
Answer generated using the input document and the synthesis
rules of the language