Research

I formulate novel algorithm based on machine learning and deep learning approaches to address emerging privacy and security issues in online and mobile environments.

My Ph.D. work in the CSE department at UB is on computer privacy and security, robustness of deep learning models, mitigating online offensiveness and studying AI bias.

I focus specifically in these areas:

• Online offensiveness—Studying emerging issues such as cyberbullying and cyberharassment, understadning their factors and mitigating them using AI
• Online hate speech—Studying hate speech and hateful memes
• Robustness of deep learning models—Studying robustness of deep multimodal models (i.e., models based on multiple modalities) against adversarial attacks
• Image privacy and security—We are researching in AI-based multiparty access control and recommender systems for private images in online social networks
• AI Bias—We are studying bias in hate speech detection AI. Currently, we are looking at how different categories exist in hate speech and how differences in these categories could induce bias in AI

Select Publications

Published Papers (* indicates high school student who worked with me)

• [CODASPY 2022] Towards Automated Content-based Photo Privacy Control in User-Centered Social Networks
  Nishant Vishwamitra, Yifang Li, Hongxin Hu, Kelly Caine, Long Cheng, Ziming Zhao and Gail-Joon Ahn
  In Proceedings of the Twelth ACM Conference on Data and Application Security and Privacy
  [Paper]
• [Journal of Hate Studies 2021] COVID-19: A Pandemic of Anti-Asian Hate
  Matthew Costello, Long Cheng, Feng Luo, Hongxin Hu, Song Liao, Nishant Vishwamitra, Mingqi Li and Ebuka Okpala
  In Journal of Hate Studies
  [Paper]


• [NDSS 2021] Towards Understanding and Detecting Cyberbullying in Real-world Images
  Nishant Vishwamitra, Hongxin Hu, Feng Luo and Long Cheng
  In Proceedings of the 28th Annual Network and Distributed System Security Symposium
  Acceptance rate: 87/573 = 15.2%
  [Paper] [Code] [Dataset]


• [ICMLA 2020] On Analyzing COVID-19-related Hate Speech Using BERT Attention
  Nishant Vishwamitra, Ruijia Hu*, Feng Luo, Long Cheng, Matthew Costello, and Yin Yang
  In Proceedings of the 19th IEEE International Conference on Machine Learning and Applications
  [Paper] [Dataset]


• [CODASPY 2020] On the Impact of Word Representation in Hate Speech and Offensive Language Detection and Explanation
  Ruijia Hu*, Wyatt Dorris*, Nishant Vishwamitra, Feng Luo and Matthew Costello
  In Proceedings of the Tenth ACM Conference on Data and Application Security and Privacy
  [Paper]


• [IWSPA 2020] Towards Automatic Detection and Explanation of Hate Speech and Offensive Language
  Wyatt Dorris*, Ruijia Hu*, Nishant Vishwamitra, Feng Luo and Matthew Costello
  In Proceedings of the Sixth International Workshop on Security and Privacy Analytics
  [Paper]


• [CHI 2020] Towards a taxonomy of content sensitivity and sharing preferences for photos
  Yifang Li, Nishant Vishwamitra, Hongxin Hu and Kelly Caine
  In Proceedings of the 2020 Conference on Human Factors in Computing Systems
  Acceptance rate: 760/3126 = 24.3%
  [Paper]


• [CSCW 2017] Effectiveness and users' experience of obfuscation as a privacy-enhancing technology for sharing photos
  Yifang Li, Nishant Vishwamitra, Bart Knijnenburg, Hongxin Hu and Kelly Caine
  In Proceedings of the 20th ACM Conference on Computer-Supported Cooperative Work and Social Computing
  Acceptance rate: 105/384 = 27.3%
  [Paper]


• [ICMLA 2018] CrescendoNet: A New Deep Convolutional Neural Network with Ensemble Behavior
  Xiang Zhang, Nishant Vishwamitra, Hongxin Hu and Feng Luo
  In Proceedings of the 17th IEEE International Conference on Machine Learning and Applications
  [Paper]


• [HFES 2017] Effectiveness and Users' Experience of Face Blurring as a Privacy-protection for Sharing Photos via Online Social Networks
  Yifang Li, Nishant Vishwamitra, Hongxin Hu, Bart Knijnenberg and Kelly Caine
  In Proceedings of Human Factors and Ergonomics Society Annual Meeting
  [Paper]


• [SACMAT 2017] Towards PII-based Multiparty Access Control for Photo Sharing in Online Social Networks
  Nishang Vishwamitra, Yifang Li, Kevin Wang, Hongxin Hu, Kelly Caine and Gail-Joon Ahn
  In Proceedings of the 22nd ACM on Symposium on Access Control Models and Technologies
  [Paper]


• [IWSPA 2017] MCDefender: Toward Effective Cyberbullying Defense in Mobile Online Social Networks
  Nishant Vishwamitra, Xiang Zhang, Jonathan Tong*, Hongxin Hu, Feng Luo, Robin Kowalski and Joseph Mazer
  In Proceedings of the 3rd ACM on International Workshop on Security And Privacy Analytics
  [Paper]


• [CVPR Workshops 2017] Blur vs. Block: Investigating the Effectiveness of Privacy-enhancing for Obfuscation Images
  Yifang Li, Nishant Vishwamitra, Bart P. Knijnenburg, Hongxin Hu and Kelly Caine
  In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops
  [Paper]


• [ICMLA 2016] Cyberbullying Detection with a Pronunciation based Convolutional Neural Network
  Xiang Zhang, Jonathan Tong*, Nishant Vishwamitra, Elizabeth Whittaker, Joseph P Mazerph, Robin Kowalski, Hongxin Hu, FengLuo, Jamie Macbeth and Edward Dillon
  In 2016 15th IEEE international conference on machine learning and applications
  [Paper]

Revision

• [MISQ] Why Do Crowds Validate False Data: Systematic Errors in Validating Crowdsourced Ground Truth During a Crisis
  Pienta D., Vishwamitra N., Berente N., Thatcher J. and S. Somanchi
  Management Information Systems Quarterly