Projects

AI for Social Good

• Detecting and Countering Abusive Language on Social Media: This project focuses on detecting abusive language on social media and countering it using NLP models.
• Social Media Analysis for Early Detection of Suicide Ideation: This project is in collaboration with the University of Ottawa and focuses on developing tools for the early detection of suicidal notes on social media.
• Multimodal Social Media Analysis for Disaster Response: This project is funded by the New Beginning Initiative Program and is in collaboration with Carleton University.
• Innovative Solutions to Improve Immigration Settlement in Canada: This project is in collaboration with the MILA institute and focuses on using AI to make immigration settlement in Canada more efficient.

Ethics in AI:

• Explainability: In this project, we use concept-based explanations and other explainability techniques to analyze a model and understand its flaws regarding safety, generalizability, fairness and robustness.
• Behavioral Analysis of Generative Models: In this project, we analyze the model’s output to uncover biases learned by text classifiers and generative models.
• Ethical Challenges in Abuisve Language Detection: In this project, we bring ethical and human rights considerations to every stage of developing an NLP system for detecting abusive language in online platforms.
• Privacy: In this project, we use NLP techniques to de-identify text.

Computational Social Science

• Uncovering and Combating Stereotypes: In this project, we use computational linguistics and machine learning techniques to uncover and combat the stereotypical views prominent in society.
• Political Ideology Detection on Social Media: This project is in collaboration with McGill University, where we use network analysis and natural language processing to understand the political ideologies expressed on social media.

Machine Learning Applications

• Medical Data Analysis: We analyzed different forms of biomedical data to increase the efficiency and accuracy of diagnosis and treatment in medical applications.
• Audio and Speech Processing: We developed neural network models for improving Farsi speech recognition.
• Facial Recognition: We developed neural network models for emotion detection and facial image generation.

Selected Papers By Research Topic

Detecting and Countering Abusive Language on Social Media:

• Svetlana Kiritchenko, Georgina Curto, Isar Nejadgholi, and Kathleen C. Fraser. (2023) Aporophobia: An Overlooked Type of Toxic Language Targeting the Poor. In Proceedings of the 7th Workshop on Online Abuse and Harms (WOAH), Toronto, ON, Canada, July 2023. [pdf]
• Isar Nejadgholi, Svetlana Kiritchenko, Kathleen C. Fraser, and Esma Balkir. (2023) Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers. In Proceedings of the 7th Workshop on Online Abuse and Harms (WOAH), Toronto, ON, Canada, July 2023. [pdf] [code]
• Isar Nejadgholi, Esma Balkir, Kathleen C. Fraser, and Svetlana Kiritchenko. (2022) Towards Procedural Fairness: Uncovering Biases in How a Toxic Language Classifier Uses Sentiment Information. In Proceedings of the Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP), Abu Dhabi, United Arab Emirates, Dec. 2022. [pdf] [code]
• Isar Nejadgholi, Kathleen C. Fraser, and Svetlana Kiritchenko. (2022). Improving Generalizability in Implicitly Abusive Language Detection with Concept Activation Vectors. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, Dublin, Ireland, May 2022. [pdf] [code]
• Esma Balkir, Isar Nejadgholi, Kathleen C. Fraser, and Svetlana Kiritchenko. (2022). Necessity and Sufficiency for Explaining Text Classifiers: A Case Study in Hate Speech Detection. In Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), Seattle, WA, USA, July 2022. [pdf] [code]
• Gunasekara, I., & Nejadgholi, I. (2018, October). A review of standard text classification practices for multi-label toxicity identification of online content. In Proceedings of the 2nd workshop on abusive language online (ALW2), 21-25. [paper]

Social Media Analysis for Early Detection of Suicide Ideation:

• Ghanadian, H., Nejadgholi, I., & Osman, H. A. (2024). Socially Aware Synthetic Data Generation for Suicidal Ideation Detection Using Large Language Models. IEEE Access.[Paper]

• Ghanadian, H., Nejadgholi, I., & Osman, H. A. (2023). ChatGPT for Suicide Risk Assessment on Social Media: Quantitative Evaluation of Model Performance, Potentials and Limitations. 13th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis (WASSA at ACL 2023), Toronto, Canada, July 2023. [Paper]

Innovative Solutions to Improve the Immigration System in Canada:

• Isar Nejadgholi, Renaud Bougueng, and Samuel Witherspoon (2017). A Semi-Supervised Training Method for Semantic Search of Legal Facts in Canadian Immigration Cases. In Legal Knowledge and Information Systems (pp. 125-134). IOS Press. [Paper]

Explainability:

• Isar Nejadgholi, Svetlana Kiritchenko, Kathleen C. Fraser, and Esma Balkir. (2023) Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers. In Proceedings of the 7th Workshop on Online Abuse and Harms (WOAH), Toronto, ON, Canada, July 2023. [pdf] [code]
• Isar Nejadgholi, Esma Balkir, Kathleen C. Fraser, and Svetlana Kiritchenko. (2022) Towards Procedural Fairness: Uncovering Biases in How a Toxic Language Classifier Uses Sentiment Information. In Proceedings of the Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP), Abu Dhabi, United Arab Emirates, Dec. 2022. [pdf] [code]
• Isar Nejadgholi, Kathleen C. Fraser, and Svetlana Kiritchenko. (2022). Improving Generalizability in Implicitly Abusive Language Detection with Concept Activation Vectors. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, Dublin, Ireland, May 2022. [pdf] [code]
• Dawkins, H., Nejadgholi, I. (2022). Region-dependent temperature scaling for certainty calibration and application to class-imbalanced token classification. Accepted for publication in Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics.[paper]
• Esma Balkir, Isar Nejadgholi, Kathleen C. Fraser, and Svetlana Kiritchenko. (2022). Necessity and Sufficiency for Explaining Text Classifiers: A Case Study in Hate Speech Detection. In Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), Seattle, WA, USA, July 2022. [pdf] [code]
• Esma Balkir, Svetlana Kiritchenko, Isar Nejadgholi, and Kathleen C. Fraser. (2022) Challenges in Applying Explainability Methods to Improve the Fairness of NLP Models. In Proceedings of the Second Workshop on Trustworthy Natural Language Processing (TrustNLP @ NAACL), Seattle, WA, USA, July 2022. [pdf]

Behavioral Analysis of Generative Models:

• Kathleen C. Fraser, Svetlana Kiritchenko, and Isar Nejadgholi (2023) Diversity is Not a One-Way Street: Pilot Study on Ethical Interventions for Racial Bias in Text-to-Image Systems. In Proceedings of the 14th International Conference on Computational Creativity (ICCC), Waterloo, ON, Canada, June 2023. [pdf]
• Kathleen C. Fraser, Isar Nejadgholi, and Svetlana Kiritchenko (2023) A Friendly Face: Do Text-to-Image Systems Rely on Stereotypes when the Input is Under-Specified? In Proceedings of the Creative AI Across Modalities Workshop (CreativeAI @ AAAI), Washington, DC, USA, Feb. 2023. [pdf]

Ethical Challenges in Abuisve Language Detection:

• Svetlana Kiritchenko, Isar Nejadgholi, and Kathleen C. Fraser. Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective. Journal of Artificial Intelligence Research, 71: 431-478, July 2021. [pdf] [Extended Abstract]
• Isar Nejadgholi and Svetlana Kiritchenko. On Cross-Dataset Generalization in Automatic Detection of Online Abuse. In Proceedings of the 4th Workshop on Online Abuse and Harms at EMNLP-2020, November 2020. [pdf]
• Svetlana Kiritchenko and Isar Nejadgholi. Towards Ethics by Design in Online Abusive Content Detection. NRC Technical Report, October 2020. [pdf]

Uncovering and Combating Stereotypes:

• Kathleen C. Fraser, Svetlana Kiritchenko, Isar Nejadgholi, and Anna Kerkhof (2023) What Makes a Good Counter-Stereotype? Evaluating Strategies for Automated Responses to Stereotypical Text. In Proceedings of the First Workshop on Social Influence in Conversations (SICon), Toronto, ON, Canada, July 2023. [pdf]
• Kathleen C. Fraser, Svetlana Kiritchenko, and Isar Nejadgholi. (2022). Computational Modelling of Stereotype Content in Text. Frontiers in Artificial Intelligence, April, 2022. [paper]
• Kathleen C. Fraser, Isar Nejadgholi, and Svetlana Kiritchenko (2021). Understanding and Countering Stereotypes: A Computational Approach to the Stereotype Content Model. In Proceedings of the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021), August 2021. [pdf]
• Kathleen C. Fraser, Svetlana Kiritchenko, and Isar Nejadgholi. (2022). Extracting Age-Related Stereotypes from Social Media Texts. In Proceedings of the Language Resources and Evaluation Conference (LREC-2022), Marseille, France, June 2022. [pdf] [project webpage]

Privacy:

• Hathurusinghe, R., Nejadgholi, I., & Bolic, M. (2021, June). A Privacy-Preserving Approach to Extraction of Personal Information through Automatic Annotation and Federated Learning. In Proceedings of the Third Workshop on Privacy in Natural Language Processing (pp. 36-45). [Paper]
• Contributors, M. (2022). BLOOM: A 176B-Parameter Open-Access Multilingual Language Model. arXiv preprint arXiv:2211.05100.[Paper]

Medical Data Analysis:

• Nejadgholi, I., Fraser, K. C., & De Bruijn, B. (2020). Extensive Error Analysis and a Learning-Based Evaluation of Medical Entity Recognition Systems to Approximate User Experience. In Proceedings of the 19th SIGBioMed Workshop on Biomedical Language Processing, 177-186.
• Nejadgholi, I., Fraser, K. C., De Bruijn, B., Li, M., LaPlante, A., & El Abidine, K. Z. (2019). Recognizing UMLS semantic types with deep learning. In Proceedings of the Tenth International Workshop on Health Text Mining and Information Analysis (LOUHI 2019), 157-167.
• Nejadgholi, I., Sadreazami, H., Baird, Z., Rajan, S., & Bolic, M. (2019). Estimation of breathing rate with confidence interval using single-channel CW radar. Journal of Healthcare Engineering, 1-15.
• Nejadgholi, I., Sadreazami, H., Rajan, S., & Bolic, M. (2019). Classification of Doppler radar reflections as preprocessing for breathing rate monitoring. IET Signal Processing, 13(1), 21-28.
• Nejadgholi, I., Rajan, S., & Bolic, M. (2016). Time-frequency based contactless estimation of vital signs of human while walking using PMCW radar. In 2016 IEEE 18th International Conference on e-Health Networking, Applications and Services (Healthcom).1-6.
• Nejadgholi, I., Caytak, H., & Bolic, M. (2016). Using bioimpedance spectroscopy parameters as real-time feedback during tDCS. In 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 5246-5249.
• Pezeshki, Z., Tafazzoli-Shadpour, M., Nejadgholi, I., Mansourian, A., & Rahbar, M. (2016). Model of cholera forecasting using artificial neural network in Chabahar City, Iran. International Journal of Enteric Pathogens, 4(1), 1-8.
• Nejadgholi, I., Davidson, T., Blais, C., Tremblay, F., & Bolic, M. (2015). Classification of responders versus non-responders to tDCS by analyzing voltage between anode and cathode during treatment session. In Proceedings of World Congress on Medical Physics and Biomedical Engineering, 990-993. ** Won the merit prize for the distinguished paper
• Caytak, H., Nejadgholi, I., Batkin, I., & Bolic, M. (2015). Bioimpedance spectroscopy method for investigating changes to intracranial dose during transcranial direct current stimulation. In 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 3448-3451.
• Nejadgholi, I., & Bolic, M. (2015). A comparative study of PCA, SIMCA and Cole model for classification of bioimpedance spectroscopy measurements. Computers in biology and medicine, 63, 42-51.
• Nejadgholi, I., Caytak, H., Bolic, M., Batkin, I., & Shirmohammadi, S. (2015). Preprocessing and parameterizing bioimpedance spectroscopy measurements by singular value decomposition. Physiological measurement, 36(5), 983-1001.
• Nejadgholi, I., Moradi, M. H., & Abdol Ali, F. (2011). Patient Independent Heart Beat Classification. Iranian Journal of Biomedical Engineering, 4(4), 279-292. (In Persian)
• Nejadgholi, I., Moradi, M. H., & Abdolali, F. (2011). Using phase space reconstruction for patient independent heartbeat classification in comparison with some benchmark methods. Computers in biology and medicine, 41(6), 411-419.
• Pezeshki, Z., Tafazzoli-Shadpour, M., Mansourian, A., Eshrati, B., Omidi, E., & Nejadqoli, I. (2012). Model of cholera dissemination using geographic information systems and fuzzy clustering means: case study, Chabahar, Iran. Public health, 126(10), 881-887.

Audio and Speech Processing:

• Dehyadegary, L., Seyyedsalehi, S. A., & Nejadgholi, I. (2011). Nonlinear enhancement of noisy speech, using continuous attractor dynamics formed in recurrent neural networks. Neurocomputing, 74(17), 2716-2724. [Paper]
• Nejadgholi, I., & Seyyedsalehi, S. A. (2009). Nonlinear normalization of input patterns to speaker variability in speech recognition neural networks. Neural computing and applications, 18, 45-55. [Paper]

Facial Recognition:

• Nejadgholi, I., SeyyedSalehi, S. A., & Chartier, S. (2017). A brain-inspired method of facial expression generation using chaotic feature extracting bidirectional associative memory. Neural Processing Letters, 46(3), 943-960.
• Nejadgholi, I., & Seyyedsalehi, S. A. (2010). Chaotic control of deterministic variability in a BAM-inspired model of memory. In Proceedings of the 10th International Conference on Intelligent Systems Design and Applications, 203-208. IEEE.
• Nejadgholi, I., Seyyedsalehi, S. A., & Chartier, S. (2012). A chaotic feature extracting BAM and its application in implementing memory search. Neural processing letters, 36(1), 69-99.
• Nejadgholi, I., & Seyyedsalehi, S. A. (2012). A new brain-inspired robust face recognition through elimination of variation features. Procedia-Social and Behavioral Sciences, 32, 204-212.
• Nejadgholi, I., Chartier, S., & Seyyedsalehi, S. A. (2013). Controlling deterministic output variability in a feature extracting chaotic BAM. Neurocomputing, 120, 298-309.