Natural Language Processing

Dr. Rajakrishnan Rajkumar’s primary research involves cognitive modelling of production and comprehension in natural language using techniques and tools from the field of Natural Language Processing (NLP). In the past he has worked in the fields of Natural Language Generation (NLG) and Machine Translation (MT).

Dr. Rajakrishnan Rajkumar and Dr. Kushal Shah are jointly working on developing a Machine Learning based algorithm to distinguish between fiction and non-fiction genres with potential applications to publishing houses and news readers. Reputed publishing houses usually receive lot of article/book submissions from budding writers, and it is not easy for human editors to read all these submissions thoroughly and pick up the good ones. Their algorithm can assist these editors in automatically identifying well-written fictional texts from amongst all submissions, thereby considerably reducing their workload. From the news perspective, their algorithm can automatically identify articles which tend to use manipulative writing styles, thereby helping the readers in being cautious while reading them.

Dr. Arpit Sharma works on applying Data Mining and NLP techniques to support the development of Cyber Physical Systems. He is currently investigating the use of Word Embeddings for detecting and identifying domain specific ambiguities in requirements document for Cyber Physical Systems. He is also interested in using the domain specific information for automatic extraction and clustering of glossary items and requirements in large industrial requirements document. Additionally, he also plans to investigate the use of NLP techniques for providing explanations of the results of data driven technologies, e.g., Machine Learning.