PUBLICATIONS

Google Scholar

Doctoral Research

True and Fair: Robust and Unbiased Fake News Detection via Interpretable Machine Learning.
Chahat Raj , Anjishnu Mukherjee and Ziwei Zhu (AAAI/ACM AIES 2023)

   The dissemination of information, and consequently, misinformation, occurs at an unprecedented speed, making it increasingly difficult to discern the credibility of rapidly circulating news. Advanced large-scale language models have facilitated the development of classifiers capable of effectively identifying misinformation. Nevertheless, these models are intrinsically susceptible to biases that may be introduced through numerous ways, including contaminated data sources or unfair training methodologies. When trained on biased data, machine learning models may inadvertently learn and reinforce these biases, leading to reduced generalization performance. This situation consequently results in an inherent "unfairness" within the system. Interpretability, referring to the ability to understand and explain the decision-making process of a model, can be used as a tool to explain these biases. Our research aims to identify the root causes of these biases in fake news detection and mitigate their presence using interpretability. We also perform inference time attacks to fairness to validate robustness.

@article{raj2023truenfair,
     title = "True and Fair: Robust and Unbiased Fake News Detection via Interpretable Machine Learning",
     author = "Raj, Chahat and Mukherjee, Anjishnu and Zhu, Ziwei",
     year = "2023",
     journal = "AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society",
}

SALSA: Salience-based Switching Attack for Adversarial Perturbations in Fake News Detection Models.
Chahat Raj , Anjishnu Mukherjee, Hemant Purohit, Antonios Anastasopoulos, and Ziwei Zhu (ECIR 2024)

   Despite advances in fake news detection algorithms, recent research reveals that machine learning-based fake news detection models are still vulnerable to carefully crafted adversarial attacks. In this landscape, traditional methods, often relying on text perturbations or heuristic-based approaches, have proven insufficient, revealing a critical need for more nuanced and context-aware strategies to enhance the robustness of fake news detection. Our research identifies and addresses three critical areas: creating subtle perturbations, preserving core information while modifying sentence structure, and incorporating inherent interpretability. We propose SALSA, an adversarial Salience-based Switching Attack strategy that harnesses salient words, using similarity-based switching to address the shortcomings of traditional adversarial attack methods. Using SALSA, we perform a two-way attack: misclassifying real news as fake and fake news as real. Due to the absence of standardized metrics to evaluate adversarial attacks in fake news detection, we further propose three new evaluation metrics to gauge the attack’s success. Finally, we validate the transferability of our proposed attack strategy across attacker and victim models, demonstrating our approach’s broad applicability and potency.

@article{raj2024salsa,
     title = "SALSA: Salience-based Switching Attack for Adversarial Perturbations in Fake News Detection Models",
     author = "Raj, Chahat and Mukherjee, Anjishnu and, Purohit, Hemant and Anastasopoulos, Antonios, and Zhu, Ziwei",
     year = "2024",
     journal = "European Conference on Information Retrieval 2024",
}

Master's Research

The Effectiveness of Social Media Engagement Strategy on Disaster Fundraising.
Vivek Velivela, Chahat Raj, Mahendra Samarawickrama and Mukesh Prasad (ISCRAM 2022)

   Social media has been a powerful tool and integral part of communication, especially during natural disasters. Social media platforms help nonprofits in effective disaster management by disseminating crucial information to various communities at the earliest. Besides spreading information to every corner of the world, various platforms incorporate many features that give access to host online fundraising events, process online donations, etc. The current literature lacks the theoretical structure investigating the correlation between social media engagement and crisis management. Large nonprofit organisations like the Australian Red Cross have upscaled their operations to help nearly 6,000 bushfire survivors through various grants and helped 21,563 people with psychological support and other assistance through their recovery program (Australian Red Cross, 2021). This paper considers the case of bushfires in Australia 2019-2020 to inspect the role of social media in escalating fundraising via analysing the donation data of the Australian Red Cross from October 2019 - March 2020 and analysing the level of public interaction with their Facebook page and its content in the same period.

@article{velivela2023effectiveness,
     title = "The Effectiveness of Social Media Engagement Strategy on Disaster Fundraising",
     author = "Velivela, Vivek and Raj, Chahat and Samarawickrama, Mahendra and Prasad, Mukesh",
     year = "2023",
     note = "ISCRAM 2022",
}

Is Dynamic Rumor Detection on Social Media Viable? An Unsupervised Perspective.
Chahat Raj and Priyanka Meel (Under Review | Arxiv)

   This paper investigates the viability of dynamic rumor detection on social media from an unsupervised perspective.

@article{raj2023dynamic,
     title = "Is Dynamic Rumor Detection on Social Media Viable? An Unsupervised Perspective",
     author = "Raj, Chahat and Meel, Priyanka",
     year = "Under Review",
     note = "Arxiv",
}

People Lie, Actions Don’t! Modeling Infodemic Proliferation Predictors among Social Media Users.
Chahat Raj and Priyanka Meel (TECHNOLOGY IN SOCIETY, ELSEVIER, IF: 7.82)

   This paper proposes a model to predict infodemic proliferation among social media users by focusing on their actions rather than their stated opinions.

@article{raj2022people,
     title = "People Lie, Actions Don’t! Modeling Infodemic Proliferation Predictors among Social Media Users",
     author = "Raj, Chahat and Meel, Priyanka",
     journal = "Technology in Society",
     volume = "70",
     pages = "101914",
     year = "2022",
     publisher = "Elsevier",
     doi = "10.1016/j.techsoc.2022.101914",
}

Cyberbullying Detection: Hybrid Models based on Machine Learning and Natural Language Processing Techniques.
Chahat Raj, Ayush Agarwal, Gnana Bharathy, Bhuva Narayan and Mukesh Prasad (ELECTRONICS, MDPI, IF: 2.397)

   This paper proposes hybrid models for cyberbullying detection, using machine learning and natural language processing techniques.

@article{raj2021cyberbullying,
     title = "Cyberbullying Detection: Hybrid Models based on Machine Learning and Natural Language Processing Techniques",
     author = "Raj, Chahat and Agarwal, Ayush and Bharathy, Gnana and Narayan, Bhuva and Prasad, Mukesh",
     journal = "Electronics",
     volume = "10",
     issue = "16",
     pages = "1947",
     year = "2021",
     publisher = "MDPI",
     doi = "10.3390/electronics10161947",
     url = "https://doi.org/10.3390/electronics10161947",
}

Machine Learning Techniques for Differential Diagnosis of Vertigo and Dizziness: A Review.
Varad Kabade, Ritika Hooda, Chahat Raj, Zainab Awan, Allison Young, Miriam Welgampola and Mukesh Prasad (SENSORS, IF: 3.576)

   This review discusses the use of machine learning techniques for the differential diagnosis of vertigo and dizziness.

@article{kabade2021machine,
     title = "Machine Learning Techniques for Differential Diagnosis of Vertigo and Dizziness: A Review",
     author = "Kabade, Varad and Hooda, Ritika and Raj, Chahat and Awan, Zainab and Young, Allison and Welgampola, Miriam and Prasad, Mukesh",
     journal = "Sensors",
     volume = "21",
     issue = "18",
     pages = "6266",
     year = "2021",
     publisher = "MDPI",
     doi = "10.3390/s21186266",
     url = "https://doi.org/10.3390/s21186266",
}

ARCNN Framework for Multimodal Infodemic Detection.
Chahat Raj and Priyanka Meel (NEURAL NETWORKS, ELSEVIER, IF: 9.657)

   This paper proposes an ARCNN framework for multimodal infodemic detection using deep learning techniques.

@article{raj2021arcnn,
     title = "ARCNN Framework for Multimodal Infodemic Detection",
     author = "Raj, Chahat and Meel, Priyanka",
     journal = "Neural Networks",
     volume = "142",
     pages = "270-279",
     year = "2021",
     publisher = "Elsevier",
     doi = "10.1016/j.neunet.2021.05.008",
     url = "https://doi.org/10.1016/j.neunet.2021.05.008",
}

Bitcoin Price Prediction using LSTM Autoencoder and False Nearest Neighbor Loss.
Chahat Raj and Manojit Chattopadhyay

   This paper proposes a deep learning approach using LSTM autoencoder and false nearest neighbor loss for Bitcoin price prediction.

@inproceedings{raj2022bitcoin,
     title = "Bitcoin Price Prediction using LSTM Autoencoder and False Nearest Neighbor Loss",
     author = "Raj, Chahat and Chattopadhyay, Manojit",
     booktitle = "Proceedings of the 2022 International Conference on Machine Learning and Data Mining",
     pages = "175-181",
     year = "2022",
     organization = "ACM",
     doi = "10.1145/3462764.3462780",
     url = "https://doi.org/10.1145/3462764.3462780",
}

Microblogs Deception Detection using BERT and Multiscale CNN
Chahat Raj and Priyanka Meel (GCAT 2021, IEEE XPLORE)

   This paper proposes a deep learning approach using BERT and multiscale CNN for deception detection in microblogs.

@inproceedings{raj2021microblogs,
     title = "Microblogs Deception Detection using BERT and Multiscale CNN",
     author = "Raj, Chahat and Meel, Priyanka",
     booktitle = "Proceedings of the 2021 Global Conference on Artificial Intelligence and Technology",
     pages = "117-123",
     year = "2021",
     organization = "IEEE",
     doi = "10.1109/GCAT51812.2021.9399382",
     url = "https://doi.org/10.1109/GCAT51812.2021.9399382",
}

MediaEval 2020: An Ensemble-based Multimodal Approach for Coronavirus and 5G Conspiracy Tweet Detection.
Chahat Raj and Mihir P Mehta (CEUR WORKSHOP PROCEEDINGS, MEDIAEVAL 2020)

   This paper presents an ensemble-based multimodal approach for detecting coronavirus and 5G conspiracy tweets in social media.

@inproceedings{raj2020mediaeval,
     title = "MediaEval 2020: An Ensemble-based Multimodal Approach for Coronavirus and 5G Conspiracy Tweet Detection",
     author = "Raj, Chahat and Mehta, Mihir P",
     booktitle = "Working Notes Proceedings of the MediaEval 2020 Workshop",
     year = "2020",
     url = "http://ceur-ws.org/Vol-2659/Mediaeval_2020_paper_23.pdf",
}

ConvNet Frameworks for Multimodal Fake News Detection
Chahat Raj and Priyanka Meel (APPLIED INTELLIGENCE, SPRINGER, IF: 5.086)

   This paper presents ConvNet frameworks for multimodal fake news detection.

@article{raj2021convnet,
     title = "ConvNet Frameworks for Multimodal Fake News Detection",
     author = "Raj, Chahat and Meel, Priyanka",
     journal = "Applied Intelligence",
     year = "2021",
     volume = "51",
     number = "2",
     pages = "898-919",
     issn = "0924-669X",
     doi = "10.1007/s10489-020-01828-4",
     url = "https://doi.org/10.1007/s10489-020-01828-4",
     publisher = "Springer US",
}

A Review of Web Infodemic Analysis and Detection Trends across Multi‑modalities using Deep Neural Networks.
Chahat Raj and Priyanka Meel (Under Review | Arxiv)

   This paper reviews trends in web infodemic analysis and detection across multimodalities using deep neural networks.

@article{raj2021web,
     title = "A Review of Web Infodemic Analysis and Detection Trends across Multi‑modalities using Deep Neural Networks",
     author = "Raj, Chahat and Meel, Priyanka",
     journal = "Neurocomputing",
     year = "2021",
     note = "Under Review",
}

Fake News on Multiple Online Social Networks.
Chahat Raj and Priyanka Meel (ICPCCAI 2020)

   This paper presents a study on fake news on multiple online social networks.

@inproceedings{raj2020fake,
     title={Fake News on Multiple Online Social Networks},
     author={Raj, Chahat and Meel, Priyanka},
     booktitle={Proceedings of the International Conference on Pervasive Computing and Communication},
     year={2020},
}

PUBLICATIONS

Doctoral Research

Master's Research

Chahat Raj