Explain your education career and current studies…

I completed my B.E in ECE from College of Engineering, Guindy in the year 1982. I then joined as Teaching Research Fellow in the department of ECE at the same institute and meanwhile also completed my M.E in Computer Science and Engineering. Later I joined the Department of Computer Science and Engineering when it was first established. I completed my Ph.D in the area of Natural Language Processing in the year 1992. I have successfully guided ten Ph.D students and I am currently guiding eleven students.

My current area of interests include Natural Language Processing, Music Processing, E-learning, Ontology based text mining, Semantic processing and Search engines.

What are the areas you are interested in NLP?

I am really interested in the area of Natural language Processing and some of my work in this area include:
• Information Visualization
• Automatic Content generation for E-learning
• Automatic Question generation for E-learning
• Automatic Slide Generation for E-learning
• Research Area Mining
• Ontology based Text Mining

Natural language Processing Vs Indic Computing?

The special characteristics of Tamil – morphological richness and relative free word order require different approaches to tackle many of the language processing tasks. We concentrate on both rule based and statistical based machine learning approaches to handle morphological analysis, semantic interpretation using UNL ( Universal networking Language – a language independent semantic representation), word sense disambiguation, anaphora resolution etc..

What are the Tamil Computing activities undergoing in Anna University ?

The Department of Computer Science and Engineering and Department of Information Science and Technology have jointly set up the “Tacola” ( Tamil Computing Lab) at Anna University, Chennai. The Tamil Computing activities at Anna University are carried out under the leadership of myself, Dr.Ranjani Parthasarathy and Dr.Madhan Karky. At present about 25 project staff work in the Tacola Lab. The work carried out include:

• The development of CoRee – The concept based Tamil Search engine
• Agaraadhi – a Tamil Dictionary with many unique features including web based analytics
• Tamil Word based Games – Web based
• Kuralagam – under development
• We have special interest in Tamil Computing and some research work being carried out in this area include
• Web based Tamil Question Answering
• Automatic Tamil document Summarization
• CoRee – Concept based Search Engine – Tamil
 Document Processing
 Indexing
 Searching and Ranking
 Summarizing
• Event Based Search Engine-Tamil
• Semantic Interpretation of Tamil Documents
• Tamil Lyric Mining

What are the Indic Computing Projects and activities undergoing now?

Cross-lingual Information Access – a Ministry of Information and Communication, Govt. of India. This project is carried out under a consortium with 7 other member institutes from all over India. We are carrying out the concept based search engine for Tamil. The Department has also been identified for offering an M.E. Programme in Knowledge Engineering and Computational Linguistics for which we designed the curriculum and syllabi.

What are the tools required for our language computing development?

Tools required for Indic computing development are basic language processing tools such as morphological analyzer, a good dictionary, and other tools for handling language input and output. Tagged data is needed for many statistical processing tasks. Good standard Corpus (like Reuters for English) needs to be made available. A good set of parallel corpora would make machine translation among Indian languages possible.

What you want to say to the Students about the scope of Indic Computing?

Students are showing tremendous interest in the area of Tamil computing. In fact a large number of students (average of 30) take up summer projects in the area. Every year at least 15 projects are in the area of Tamil Computing. This summer a number of students took up mobile app development for Tamil Computing Applications. This is an area with lot of opportunities for students. Exposure to this type of language computing also enables students to take up jobs in areas requiring text mining which is just picking up.

What you want to say to |MNC’s ?

Corporates and MNCs need to harness the students trained in this challenging area and create applications to cater to the vast majority of people with only knowledge of local language. The development of speech enabled applications for this group would also be a challenging task.