Language Technology Lab | Piush Aggarwal

Piush Aggarwal
PhD Candidate
piush.aggarwal@fernuni-hagen.de

Universitätsstraße 27 – PRG / Gebäude 5, Raum B117 / 1. OG, 58097 Hagen

As a computer scientist in NLP, I work on analysing robustness and explanations of multimodal systems having language as one of the modality.

Research Interests

Multimodal Solutions to NLP applications
Robustness of ML systems
Explanation AI

Short Vita

From April 2022: Research assistant at professorship “Computational Linguistic” at the Research Cluster D²L² (Digitalization, Diversity and Lifelong Learning), FernUniversität in Hagen
01.2019 - 03.2022: Researcher at Language Technology Lab, Universität Duisburg-Essen
04.2017 - 07.2019: M.Sc. in Computer Engineering with Specialization in Intelligent Networked Systems at University of Duisburg-Essen
11.2017 - 10.2018: Human Language Technology’s Assistant at eBay Inc.
09.2015 - 03.2017: Business Intelligence Engineer at Bank of America
10.2014 - 09.2015: e-Discovery Engineer at Xerox
06.2012 - 10.2014: Business Intelligence Engineer at Standard Chartered Bank
08.2008 – 03.2012: B.Tech. in Electronics and Communications at Guru Gobind Singh Indraprastha University, Delhi, India

Publications

2022

2021

VL-BERT+: Detecting Protected Groups in Hateful Multimodal Memes

Aggarwal, Piush, Liman, Michelle Espranita, Gold, Darina, and Zesch, Torsten

In Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021) Aug 2021

Abstract PDF

This paper describes our submission (winning solution for Task A) to the Shared Task on Hateful Meme Detection at WOAH 2021. We build our system on top of a state-of-the-art system for binary hateful meme classification that already uses image tags such as race, gender, and web entities. We add further metadata such as emotions and experiment with data augmentation techniques, as hateful instances are underrepresented in the data set.
Fully vs. Weakly Supervised Caries Localization in Smartphone Images with CNNs

Pham, Duc Duy, Müller, Jonas, Aggarwal, Piush, Khatri, Amit, Sharma, Mayank, Zesch, Torsten, and Pauli, Josef

In Artificial Intelligence for Healthcare Applications International Workshop - ICPR 2020 Workshop Proceedings Jan 2021

2019

Identification of Good and Bad News on Twitter

Aggarwal, Piush, and Aker, Ahmet

In Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019) Sep 2019

Abstract PDF

Social media plays a great role in news dissemination which includes good and bad news. However, studies show that news, in general, has a significant impact on our mental stature and that this influence is more in bad news. An ideal situation would be that we have a tool that can help to filter out the type of news we do not want to consume. In this paper, we provide the basis for such a tool. In our work, we focus on Twitter. We release a manually annotated dataset containing 6,853 tweets from 5 different topical categories. Each tweet is annotated with good and bad labels. We also investigate various machine learning systems and features and evaluate their performance on the newly generated dataset. We also perform a comparative analysis with sentiments showing that sentiment alone is not enough to distinguish between good and bad news.
Classification Approaches to Identify Informative Tweets

Aggarwal, Piush

In Proceedings of the Student Research Workshop Associated with RANLP 2019 Sep 2019

Abstract PDF

Social media platforms have become prime forums for reporting news, with users sharing what they saw, heard or read on social media. News from social media is potentially useful for various stakeholders including aid organizations, news agencies, and individuals. However, social media also contains a vast amount of non-news content. For users to be able to draw on benefits from news reported on social media it is necessary to reliably identify news content and differentiate it from non-news. In this paper, we tackle the challenge of classifying a social post as news or not. To this end, we provide a new manually annotated dataset containing 2,992 tweets from 5 different topical categories. Unlike earlier datasets, it includes postings posted by personal users who do not promote a business or a product and are not affiliated with any organization. We also investigate various baseline systems and evaluate their performance on the newly generated dataset. Our results show that the best classifiers are the SVM and BERT models.
LTL-UDE at SemEval-2019 Task 6: BERT and Two-Vote Classification for Categorizing Offensiveness

Aggarwal, Piush, Horsmann, Tobias, Wojatzki, Michael, and Zesch, Torsten

In Proceedings of the International Workshop on Semantic Evaluation (SemEval) Sep 2019

PDF

2018

Information Nutrition Labels: A Plugin for Online News Evaluation

Kevin, Vincentius, Högden, Birte, Schwenger, Claudia, Sahan, Ali, Madan, Neelu, Aggarwal, Piush, Bangaru, Anusha, Muradov, Farid, and Aker, Ahmet

In Proceedings of the First Workshop on Fact Extraction and VERification (FEVER) Nov 2018

Abstract PDF

In this paper we present a browser plugin \textitNewsScan that assists online news readers in evaluating the quality of online content they read by providing \textitinformation nutrition labels for online news articles. In analogy to groceries, where nutrition labels help consumers make choices that they consider best for themselves, information nutrition labels tag online news articles with data that help readers judge the articles they engage with. This paper discusses the choice of the labels, their implementation and visualization.