Research Interest: My broad interests lie in Social Computing, Natural Language Understanding, and Human-AI Collaboration for robust applications in high-risk, dynamic environments at public services and nonprofits. I design interactive AI systems for real-time data analytics and information management of multi-source, multi-modal data that augment analytical capabilities while ensuring human control of algorithmic behavior. These systems rely upon human-centered computing methods that facilitate effective human-AI collaboration to observe, model, and analyze the behavior of individuals and groups in online and offline settings at scale, using new semantic computing, NLP, and ML techniques guided by socio-psychological theories.   The application of these systems includes the preparedness, response, and recovery to natural crises (e.g., hurricanes), societal crises (e.g., hate, violence), and human crises (e.g., cyber attacks, scams).   

Technical Interest: Fuse top-down and bottom-up data mining approaches to develop explainable computational models with domain knowledge and cognitive/social theories, which helps in both predicting and explaining implicit behavior in unstructured social, web, and IoT data streams at scale.

Relevant projects: Check Humanitarian Informatics Lab (Human_Info_Lab) research group page.

Grants and Acknowledgement: We are grateful to the following agencies for supporting our research and its integration into education: U.S. National Science Foundation (NSF), Commonwealth Cyber Initiative (CCI), Virginia; Oak Ridge Institute for Science and Education (ORISE), U.S. Office of Navel Research (ONR), U.S. Department of Homeland Security (DHS), 4VA Program and CAHMP Center at Mason, as well as Western Norway Research Institute with Research Council of Norway (RCN):

• SOCYTI: Violence-inducing Behaviour Prevention in Social-Cyber Space of Local Communities, Sponsor: RCN, Period: 2022-26
• EAGER: DCL: SaTC: EIC: Inclusive-ScamBuster: Inclusive Scam Detection Methods for Social Media to Design Assistive Tools for Protecting Individuals with Developmental Disabilities, Sponsor: NSF, Period: 2022-24
• Holistic Human Identity Mapping, Sponsor: ORISE, Period: 2021-23
• Characterizing Biases in Automated Scam Detection Tools for Social Media to Aid Individuals with Developmental Disabilities, Sponsor: CCI, Period: 2022
• Designing Tools for Proactive Counter-disinformation Communication to Empower Local Government Agencies, Sponsor: CCI, Period: 2021-22
• Assessing Tobacco Prevention Policies Using a Hybrid Approach of Nontraditional Social Media and Traditional Simulation Modeling, Sponsor: 4VA Mason, Period: 2021-22
• ScamBuster: Detecting Emergent Scam Networks on Social Media during Disasters, Sponsor: CAHMP Mason, Period: 2021-22
• SCC-CIVIC-PG Track B: Assessing the Feasibility of Systematizing Human-AI Teaming to Improve Community Resilience, Sponsor: NSF, Period: 2021
• INTPART: Transnational Partnership for Excellent Research & Education in Disruptive Technologies for a Resilient Future, Sponsor: RCN, Period: 2020-23
• CMMI: RAPID: Collaborative Research: Human-AI Teaming for Big Data Analytics to Enhance Response to the COVID-19 Pandemic, Sponsor: NSF, Period: 2020-21
• Research Team Commerical First Innovation TM a.k.a. Smart Community IoT Innovation (SCITI Labs), Sponsor: Center for Innovative Technology and DHS, Period: 2020
• IIS: III: Summarizing Heterogeneous Crowdsourced & Web Streams Using Uncertain Concept Graphs, Sponsor: NSF, Period: 2018-21
• CRII:CHS: Mining Intentions on Social Media to Enhance Situational Awareness of Crisis Response Organizations, Sponsor: NSF, Period: 2017-20
• EAGER: Social Media Participation as Indicator of Actors, Awareness, Attitudes, and Activities Related to STEM Education, Sponsor: NSF, Period: 2017-20
• INTPART: Transnational Partnership for Excellent Research & Education in Big Data and Emergency Management, Sponsor: RCN, Period: 2017-20
• Believable Fake Scientific Document Generation by Exploring Data Semantics, Sponsor: ONR, Period: 2016-20

Selected Publications: (full list: Google Scholar Profile)

• - Semantic computing for behavior modeling and intent mining on social media
  • F. Vitiugin, Y. Senarath, and H. Purohit. (2021). Efficient Detection of Multilingual Hate Speech by Using Interactive Attention Network with Minimal Human Feedback. 13th ACM Web Science Conference (WebSci'21), pp. 130–138, DOI: 10.1145/3447535.3462495 (Acceptance Rate: 29%)
  • J. Krishnan, H. Purohit, and H. Rangwala. (2020). Diversity-Based Generalization for Unsupervised Text Classification under Domain Shift. In Proceedings of the 19th European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD). (Acceptance Rate 19%)
  • H. Purohit, and R. Pandey. (2019). Intent Mining for the Good, Bad & Ugly Use of Social Web: Concepts, Methods, and Challenges. In Emerging Research Challenges and Opportunities in Computational Social Network Analysis and Mining, LNSN, Springer.
  • R. Pandey, H. Purohit, B. Stabile, and A. Grant. (2018). Distributional Semantics Approach to Detect Intent in Twitter Conversations on Sexual Assaults. In Proceedings of the 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI). (Acceptance Rate 30%) [*Best Paper Award*]
  • H. Purohit, C. Castillo, M. Imran, and R. Pandey. (2018). Social-EOC: Serviceability Model to Rank Social Media Requests for Emergency Operation Centers. In Proceedings of the 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM). (Acceptance Rate 15%)
  • H. Purohit, G. Dong, V. Shalin, K. Thirunarayan, and A. Sheth. (2015) Intent Classification of Short-Text on Social Media. The 8th IEEE International Conference on Social Computing and Networking. (Acceptance Rate 27%)
• - Disaster informatics using realtime social media analytics methods
  • Y. Senarath, R. Pandey, S. Peterson, and H. Purohit. (2023). Citizen-Helper System for Human-Centered AI Use in Disaster Management. International Handbook of Disaster Research, Springer, pp. 1–21.
  • Y. Senarath, H. Purohit, A. Mukhopadhyay and A. Dubey. (2024). Designing a Human-centered AI Tool to Support Emergency Response through Proactive Incident Detection from Non-traditional Data Sources. ACM Digital Government: Research and Practice, 5(1), DOI: 10.1145/3633784.
  • J. Krishnan, A. Anastasopoulos, H. Purohit and H. Rangwala. (2022). Cross-Lingual Text Classification of Transliterated Hindi and Malayalam. 2022 IEEE International Conference on Big Data (Big Data), pp. 1850-1857. (Acceptance Rate: 18.6%)
  H. Purohit, C. Castillo, and R. Pandey (2020). Ranking and grouping social media requests for emergency services using serviceability model. Social Network Analysis and Mining (SNAM), 10(1), 1-17.
  • H. Purohit, R. Kanagasabai, and N. Deshpande. (2019). Towards Next Generation Knowledge Graphs for Disaster Management. In Proceedings of the 2019 IEEE 13th International Conference on Semantic Computing (ICSC).
  • B. Pedrood, and H. Purohit. (2018). Mining Help Intent on Twitter during Disasters via Transfer Learning with Sparse Coding. In Proceedings of the 11th Int'l Conference on Social, Cultural, Behavioral Modeling (SBP-BRiMS). (Acceptance Rate 21%)
  • H. Purohit, C. Castillo, M. Imran, and R. Pandey. (2018). Ranking of Social Media Alerts with Workload Bounds in Emergency Operation Centers. In Proceedings of the 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI). (Acceptance Rate 30%)
• - Trust, influence, and engagement in dynamic communities on social media
  • F. Amro and H. Purohit. (2023).Integrated content-network analysis to discover influential collectives for studying social cyber-threats from online social movements. Social Network Analysis and Mining, vol. 13, no. 1, p. 120.
  • Z. Rajabi, A. Shehu, and H. Purohit. (2019). User Behavior Modelling for Fake Information Mitigation on Social Web . In Proceedings of the 12th Int'l Conference on Social Computing, Behavioral-Cultural Modeling and Prediction and Behavior Representation in Modeling and Simulation (SBP-BRiMS 2019).
  • H. Karbasian, H. Purohit, R. Handa, A. Malik, and A. Johri. (2018). Real-Time Inference of User Types to Assist with more Inclusive and Diverse Social Media Activism Campaigns. In Proceedings of AAAI International Conference on AI, Ethics and Society, 2018. (Acceptance Rate: 37%)
  • R. Pandey, H. Purohit, J. Chan, and A. Johri. (2018). AI for Trustworthiness! Credible User Identification on Social Web for Disaster Response Agencies. AAAI 2018 Fall Symposium Series.
  • H. Purohit, and J. Chan. (2017). Classifying User Types on Social Media to inform Who-What-Where Coordination during Crisis Response. The 14th International Conference on Information Systems for Crisis Response and Management (ISCRAM).
  • H. Purohit, T. Banerjee, A. Hampton, V. L. Shalin, N. Bhandutia, and A. P. Sheth. (2016). Gender-based violence in 140 characters or fewer: a #BigData case study of Twitter. First Monday, Vol. 21(1). (Selected Highlight of FM January'16 issue)
• - Human-AI collaboration and stream analytics systems for web intelligence and event summarization:
  • R. Pandey, H. Purohit, C. Castillo, and V.L. Shalin (2022). Modeling and mitigating human annotation errors to design efficient stream processing systems with human-in-the-loop machine learning. International Journal of Human-Computer Studies, 160, 102772.
  • R. Pandey, B. Bannan, and H. Purohit. (2020). CitizenHelper-training: AI-infused System for Multimodal Analytics to assist Training Exercise Debriefs at Emergency Services. In Proceedings of the 17th International Conference on Information Systems for Crisis Response & Management (ISCRAM) (pp. 42-53).
  • R. Pandey, C. Castillo, and H. Purohit. (2019). Modeling Human Annotation Errors to Design Bias-Aware Systems for Social Stream Processing. In Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM). (Acceptance Rate 29%)
  • R. Pandey, and H. Purohit. (2018). CitizenHelper-Adaptive: Expert-augmented Streaming Analytics System for Emergency Services and Humanitarian Organizations. In Proceedings of the 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM).
  • A. Sheth, H. Purohit, G. A. Smith, J. Brunn, A. Jadhav, P. Kapanipathi, C. Lu, and W. Wang. (2017). Twitris: A system for collective social intelligence. In R. Alhajj & J. Rokne (Eds.), Encyclopedia of Social Network Analysis and Mining, Springer.
  • P. Karuna, M. Rana, and H. Purohit. (2017). CitizenHelper: A Streaming Analytics System to Mine Citizen and Web Data for Humanitarian Organizations. The 11th International AAAI Conference on Web and Social Media (ICWSM).
• - Semantic text analytics to generate Information overload on cyber attackers for defensive cyber deception:
  • P. Karuna, H. Purohit, S. Jajodia, R. Ganesan, and O. Uzuner. (2020). Fake Document Generation for Cyber Deception by Manipulating Text Comprehensibility. IEEE Systems Journal.
  • P. Karuna, H. Purohit, R. Ganesan, and S. Jajodia. (2018). Generating Hard to Comprehend Fake Documents for Defensive Cyber Deception. IEEE Intelligent Systems.
  • P. Karuna, H. Purohit, O. Uzuner, R. Ganesan, and S. Jajodia. (2018). Enhancing Cohesion and Coherence of Fake Text to Improve Believability for Deceiving Cyber Attackers. In Language, Cognition and Computational Models workshop, COLING 2018.