• Publications
  • Influence
Deep Learning for Hate Speech Detection in Tweets
These experiments on a benchmark dataset of 16K annotated tweets show that such deep learning methods outperform state-of-the-art char/word n-gram methods by ~18 F1 points. Expand
MVAE: Multimodal Variational Autoencoder for Fake News Detection
An end-to-end network that uses a bimodal variational autoencoder coupled with a binary classifier for the task of fake news detection, which outperforms state-of-the-art methods by margins as large as ~ 6% in accuracy and ~ 5% in F1 scores. Expand
Towards Sub-Word Level Compositions for Sentiment Analysis of Hindi-English Code Mixed Text
This paper introduces a Hindi-English (Hi-En) code-mixed dataset for sentiment analysis and performs empirical analysis comparing the suitability and performance of various state-of-the-art SA methods in social media. Expand
IIIT Hyderabad at TAC 2009
This paper describes the participation at TAC 2008 in all the three tracks and introduces two major features, a feature based on Information Loss if the authors don’t pick a particular sentence and a language modeling extension that boosts novel terms and penalizes stale terms. Expand
Towards Deep Semantic Analysis of Hashtags
A context aware approach to segment and link entities in the hashtags to a knowledge base (KB) entry, based on the context within the tweet, which demonstrates the effectiveness of the technique in improving the overall entity linking in tweets via additional semantic information provided by segmenting and linking entities in a hashtag. Expand
Author Profiling: Predicting Age and Gender from Blogs Notebook for PAN at CLEF 2013
This paper proposes a Machine Learning approach to determine unknown author's age and gender using three types of features: content based, style based and topic based. Expand
Dynamic energy efficient data placement and cluster reconfiguration algorithm for MapReduce framework
This paper addresses energy conservation for clusters of nodes that run MapReduce jobs by dynamically reconfiguring the cluster based on the current workload and turns cluster nodes on or off when the average cluster utilization rises above or falls below administrator specified thresholds, respectively. Expand
How education, stimulation, and incubation encourage student entrepreneurship: Observations from MIT, IIIT, and Utrecht University
Abstract Universities across the world are increasingly trying to become more entrepreneurial, in order to stay competitive, generate new sources of income through licensing or contract research, andExpand
Network-aware virtual machine consolidation for large data centers
A greedy consolidation algorithm is proposed that ensures the number of migrations is small and the placement decisions are fast, which makes it practical for large data centers and observed a ~70% savings of the interconnect bandwidth and overall ~60% improvements in the applications performances. Expand
Deep Neural Architecture for News Recommendation
This work presents a deep neural model, where a non-linear mapping of users and item features are learnt first and a ranking based objective function is used to learn the parameters of the network. Expand