no code implementations • GWC 2018 • Kevin Patel, Pushpak Bhattacharyya
Given a word, what is the most frequent sense in which it occurs in a given corpus?
no code implementations • LREC 2018 • Diptesh Kanojia, Kevin Patel, Pushpak Bhattacharyya
Linked wordnets are extensions of wordnets, which link similar concepts in wordnets of different languages.
no code implementations • GWC 2018 • Kevin Patel, Diptesh Kanojia, Pushpak Bhattacharyya
Thus techniques that can aid the experts are desirable.
no code implementations • GWC 2019 • Diptesh Kanojia, Kevin Patel, Pushpak Bhattacharyya, Malhar Kulkarni, Gholamreza Haffari
Automatic Cognate Detection (ACD) is a challenging task which has been utilized to help NLP applications like Machine Translation, Information Retrieval and Computational Phylogenetics.
no code implementations • ACL 2018 • Sandeep Mathias, Diptesh Kanojia, Kevin Patel, Samarth Agarwal, Abhijit Mishra, Pushpak Bhattacharyya
Such subjective aspects are better handled using cognitive information.
no code implementations • COLING 2018 • Girishkumar Ponkiya, Kevin Patel, Pushpak Bhattacharyya, Girish Palshikar
It has been observed that uncovering the preposition is a significant step towards uncovering the predicate.
no code implementations • IJCNLP 2017 • Kevin Patel, Pushpak Bhattacharyya
More specifically, we show that the number of pairwise equidistant words of the corpus vocabulary (as defined by some distance/similarity metric) gives a lower bound on the the number of dimensions , and going below this bound results in degradation of quality of learned word embeddings.
no code implementations • WS 2017 • Kevin Patel, Divya Patel, Mansi Golakiya, Pushpak Bhattacharyya, Nilesh Birari
We add information from medical coding data, as well as the first level from the hierarchy of ICD-10 medical code set to different pre-trained word embeddings.
no code implementations • EMNLP 2016 • Aditya Joshi, Vaibhav Tripathi, Kevin Patel, Pushpak Bhattacharyya, Mark Carman
For example, this augmentation results in an improvement in F-score of around 4\% for three out of these four feature sets, and a minor degradation in case of the fourth, when Word2Vec embeddings are used.