Magazine article AI Magazine

A Perspective on AI Research in India

Magazine article AI Magazine

A Perspective on AI Research in India

Article excerpt

Artificial intelligence in India has been pursued by a passionate few over the last few decades. It has not been as widespread as in Europe and the USA. This could be due to two reasons. One is that research groups in general have been slow to gain in strength and have typically formed around a few diehard individuals scattered across the country. The priority in the first 50 years of independent India had been on undergraduate engineering education, and during this period students had inevitably gone westwards for doctoral studies, often staying back. The second was the propensity of the computing industry toward more lucrative assignments in the service sector. Both these factors are changing, not least because leading international software companies have set up research and development centers in the country.

Computer science education established itself in India in the early 1980s when the Indian Institutes of Technology (IITs) set up computer science departments and started offering undergraduate programs in the discipline. Research in artificial intelligence took off soon afterward when the government of India launched the Knowledge Based Computing Systems (KBCS) program in conjunction with the United Nations Development Program (Saint-Dizier 1991). A number of nodal centers were set up to focus on different areas of research including expert systems (IIT Madras), speech processing (Tata Institue of Fundamental Research), parallel processing (Indian Institute for Science), image processing (Indian Statistical Institute), and natural language processing (Center for Development of Advanced Computing).

Some of the early research in AI was motivated by societal needs. A prime example of this is the system Eklavya, a knowledge-based program designed to support a community health worker in dealing with symptoms of illness in toddlers (Chandrasekhara, Shanthi, and Mahabala 1994). The program, named after a disadvantaged character in the epic Mahabharata, was meant to help create systematic case histories, provide basic treatment advice, or indicate the need for a referral, much like the call center software deployed in more modern times. Other examples are the language teaching system Vidya, the flight scheduling expert system Sarani (developed at CDAC, Mumbai), and a speech synthesis system developed for the railways by TIFR. Some of the people working in the KBCS projects found their way into the software industry and wee instrumental in seeding in-house projects. For example Vivek Balaraman and his team at Tata Research Centre Pune developed a case-based reasoning kernel.

In the rest of the column we take a closer look at two strands of AI research in India that have thrived in the post-KBCS era.

Machine Translation

India is home to hundreds of different languages, with 22 being designated as official. Given that an average individual is familiar with only a few, machine translation (MT) and more recently cross language information retrieval has been a magnet for researchers. One group that has moved significantly beyond toy demonstrations is led by Rajeev Sangal and Vineet Chaitanya. The group, which had its genesis at IIT Kanpur in the 1980s, is currently active and growing at IIIT Hyderabad.

Fully aware that machine translation is a hard problem, the group embarked on translation with a basic system called Anusaaraka. The system exploited the fact that many Indian languages have well defined ways of depicting case markers by inflexions on words (Bharati, Chaitanya, and Sangal 1994). Coupled with the fact that these explicit markers, called vibhakti, can be mapped across languages, Anusaaraka, which means "the conformist," is designed to do a translation in which the reader actively brings to the fore her or his world knowledge to quickly get a gist of the content. The emphasis is on comprehensibility and access to content as opposed to grammatical correctness. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.