Rezaur Rashid

Jawad Chowdhury, Rezaur Rashid, Gabriel Terejanu

Causal modeling provides us with powerful counterfactual reasoning and interventional mechanism to generate predictions and reason under various what-if scenarios. However, causal discovery using observation data remains a nontrivial task due to unobserved confounding factors, finite sampling, and changes in the data distribution. These can lead to spurious cause-effect relationships. To mitigate these challenges in practice, researchers augment causal learning with known causal relations. The goal of the paper is to study the impact of expert knowledge on causal relations in the form of additional constraints used in the formulation of the nonparametric NOTEARS. We provide a comprehensive set of comparative analyses of biasing the model using different types of knowledge. We found that (i) knowledge that corrects the mistakes of the NOTEARS model can lead to statistically significant improvements, (ii) constraints on active edges have a larger positive impact on causal discovery than inactive edges, and surprisingly, (iii) the induced knowledge does not correct on average more incorrect active and/or inactive edges than expected. We also demonstrate the behavior of the model and the effectiveness of domain knowledge on a real-world dataset.

Rezaur Rashid, Jawad Chowdhury, Gabriel Terejanu

Causal structure learning from observational data remains a non-trivial task due to various factors such as finite sampling, unobserved confounding factors, and measurement errors. Constraint-based and score-based methods tend to suffer from high computational complexity due to the combinatorial nature of estimating the directed acyclic graph (DAG). Motivated by the `Cause-Effect Pair' NIPS 2013 Workshop on Causality Challenge, in this paper, we take a different approach and generate a probability distribution over all possible graphs informed by the cause-effect pair features proposed in response to the workshop challenge. The goal of the paper is to propose new methods based on this probabilistic information and compare their performance with traditional and state-of-the-art approaches. Our experiments, on both synthetic and real datasets, show that our proposed methods not only have statistically similar or better performances than some traditional approaches but also are computationally faster.

Jawad Chowdhury, Rezaur Rashid, Gabriel Terejanu

Understanding affective polarization in online discourse is crucial for evaluating the societal impact of social media interactions. This study presents a novel framework that leverages large language models (LLMs) and domain-informed heuristics to systematically analyze and quantify affective polarization in discussions on divisive topics such as climate change and gun control. Unlike most prior approaches that relied on sentiment analysis or predefined classifiers, our method integrates LLMs to extract stance, affective tone, and agreement patterns from large-scale social media discussions. We then apply a rule-based scoring system capable of quantifying affective polarization even in small conversations consisting of single interactions, based on stance alignment, emotional content, and interaction dynamics. Our analysis reveals distinct polarization patterns that are event dependent: (i) anticipation-driven polarization, where extreme polarization escalates before well-publicized events, and (ii) reactive polarization, where intense affective polarization spikes immediately after sudden, high-impact events. By combining AI-driven content annotation with domain-informed scoring, our framework offers a scalable and interpretable approach to measuring affective polarization. The source code is publicly available at: https://github.com/hasanjawad001/llm-social-media-polarization.

Soheil Hashtarkhani, Rezaur Rashid, Christopher L Brett et al.

Electronic health records contain inconsistently structured or free-text data, requiring efficient preprocessing to enable predictive health care models. Although artificial intelligence-driven natural language processing tools show promise for automating diagnosis classification, their comparative performance and clinical reliability require systematic evaluation. The aim of this study is to evaluate the performance of 4 large language models (GPT-3.5, GPT-4o, Llama 3.2, and Gemini 1.5) and BioBERT in classifying cancer diagnoses from structured and unstructured electronic health records data. We analyzed 762 unique diagnoses (326 International Classification of Diseases (ICD) code descriptions, 436free-text entries) from 3456 records of patients with cancer. Models were tested on their ability to categorize diagnoses into 14predefined categories. Two oncology experts validated classifications. BioBERT achieved the highest weighted macro F1-score for ICD codes (84.2) and matched GPT-4o in ICD code accuracy (90.8). For free-text diagnoses, GPT-4o outperformed BioBERT in weighted macro F1-score (71.8 vs 61.5) and achieved slightly higher accuracy (81.9 vs 81.6). GPT-3.5, Gemini, and Llama showed lower overall performance on both formats. Common misclassification patterns included confusion between metastasis and central nervous system tumors, as well as errors involving ambiguous or overlapping clinical terminology. Although current performance levels appear sufficient for administrative and research use, reliable clinical applications will require standardized documentation practices alongside robust human oversight for high-stakes decision-making.

Rezaur Rashid, Gabriel Terejanu

Causal discovery from observational data is challenging, especially with large datasets and complex relationships. Traditional methods often struggle with scalability and capturing global structural information. To overcome these limitations, we introduce a novel graph neural network (GNN)-based probabilistic framework that learns a probability distribution over the entire space of causal graphs, unlike methods that output a single deterministic graph. Our framework leverages a GNN that encodes both node and edge attributes into a unified graph representation, enabling the model to learn complex causal structures directly from data. The GNN model is trained on a diverse set of synthetic datasets augmented with statistical and information-theoretic measures, such as mutual information and conditional entropy, capturing both local and global data properties. We frame causal discovery as a supervised learning problem, directly predicting the entire graph structure. Our approach demonstrates superior performance, outperforming both traditional and recent non-GNN-based methods, as well as a GNN-based approach, in terms of accuracy and scalability on synthetic and real-world datasets without further training. This probabilistic framework significantly improves causal structure learning, with broad implications for decision-making and scientific discovery across various fields.

Gabriel Terejanu, Jawad Chowdhury, Rezaur Rashid et al.

The vast majority of research on explainability focuses on post-explainability rather than explainable modeling. Namely, an explanation model is derived to explain a complex black box model built with the sole purpose of achieving the highest performance possible. In part, this trend might be driven by the misconception that there is a trade-off between explainability and accuracy. Furthermore, the consequential work on Shapely values, grounded in game theory, has also contributed to a new wave of post-explainability research on better approximations for various machine learning models, including deep learning models. We propose a new architecture that inherently produces explainable predictions in the form of additive feature attributions. Our approach learns a graph representation for each record in the dataset. Attribute centric features are then derived from the graph and fed into a contribution deep set model to produce the final predictions. We show that our explainable model attains the same level of performance as black box models. Finally, we provide an augmented model training approach that leverages the missingness property and yields high levels of consistency (as required for the Shapely values) without loss of accuracy.