Home

I am a PhD candidate at the Department of CSE, IIT Bombay, since Autumn 2014. I am working under the guidance of Prof. Pushpak Bhattacharyya and Prof. Shivaram Kalyanakrishnan.

Research Interests

The Zero'th-law of machine learning says Garbage In Garbage Out. No matter how good your learning algorithm is, if the data (or its representation) is not good, one cannot expect the system to learn properly.

Keeping this in mind, I am currently pursuing studies in word embeddings. Word embeddings are distributed representations of words such that the cosine similarities between word embeddings correlates with the semantic similarities among the corresponding words.They are the defacto representations for words in modern deep learning systems for Natural Language Processing.

More about my research can be found here.

Research

I am currently pursuing studies in word embeddings. Word embeddings are distributed representations of words such that the cosine similarities between word embeddings correlates with the semantic similarities among the corresponding words. I am primarily invested in

• • Properties of word embeddings
• • Different hyperparameters and their effects on performance of word embeddings
• • Creating embeddings for complex units (synsets, noun compounds, etc.)
• • Studying impact of word embeddings towards model explainability

My CV

Publications

1. Towards Lower Bounds on Number of Dimensions for Word Embeddings, Kevin Patel and Pushpak Bhattcharyya, In IJCNLP'17
3. An Iterative Approach for Unsupervised Most Frequent Sense Detection using WordNet and Word Embeddings, Kevin Patel and Pushpak Bhattacharyya, In GWC'18
5. Semi-automatic WordNet Linking using Word Embeddings, Kevin Patel, Diptesh Kanojia and Pushpak Bhattacharyya, In GWC'18
7. Adapting Pre-trained Word Embeddings For Use In Medical Coding, Kevin Patel, Divya Patel, Mansi Golakiya, Pushpak Bhattacharyya and Nilesh Birari, In BioNLP'17
9. Are Word Embedding-based Features Useful for Sarcasm Detection? Aditya Joshi, Vaibhav Tripathi, Kevin Patel, Pushpak Bhattacharyya and Mark Carman, In EMNLP'16
11. Treat us like the sequences we are: Prepositional Paraphrasing of Noun Compounds using LSTM, Girishkumar Ponkiya, Kevin Patel, Pushpak Bhattacharyya and Girish Palshikar, In COLING'18
13. Eyes are the Windows to the Soul: Predicting the Rating of Text Quality Using Gaze Behaviour, Sandeep Mathias, Diptesh Kanojia, Kevin Patel, Samarth Agrawal, Abhijit Mishra and Pushpak Bhattacharyya, In ACL'18

Contact