Utkarsh Sinha


2021

pdf
CNLP-NITS @ LongSumm 2021: TextRank Variant for Generating Long Summaries
Darsh Kaushik | Abdullah Faiz Ur Rahman Khilji | Utkarsh Sinha | Partha Pakray
Proceedings of the Second Workshop on Scholarly Document Processing

The huge influx of published papers in the field of machine learning makes the task of summarization of scholarly documents vital, not just to eliminate the redundancy but also to provide a complete and satisfying crux of the content. We participated in LongSumm 2021: The 2nd Shared Task on Generating Long Summaries for scientific documents, where the task is to generate long summaries for scientific papers provided by the organizers. This paper discusses our extractive summarization approach to solve the task. We used TextRank algorithm with the BM25 score as a similarity function. Even after being a graph-based ranking algorithm that does not require any learning, TextRank produced pretty decent results with minimal compute power and time. We attained 3rd rank according to ROUGE-1 scores (0.5131 for F-measure and 0.5271 for recall) and performed decently as shown by the ROUGE-2 scores.

2020

pdf
Human Behavior Assessment using Ensemble Models
Abdullah Faiz Ur Rahman Khilji | Rituparna Khaund | Utkarsh Sinha
Proceedings of the 18th Annual Workshop of the Australasian Language Technology Association

Behavioral analysis is a pertinent step in today’s automated age. It is important to judge a statement on a variety of parameters before reaching a valid conclusion. In today’s world of technology and automation, Natural language processing tools have benefited from growing access to data in order to analyze the context and scenario. A better understanding of human behaviors would empower a range of automated tools to provide users a customized experience. For precise analysis, behavior understanding is important. We have experimented with various machine learning techniques, and have obtained a maximum private score of 0.1033 with a public score of 0.1733. The methods are described as part of the ALTA 2020 shared task. In this work, we have enlisted our results and the challenges faced to solve the problem of the human behavior assessment.