I am a researcher in Computer Science at Loyola University Chicago, working at the intersection of adversarial machine learning, natural language processing, and social computing. I build AI systems that are technically strong, socially aware, and dependable in high-stakes environments.
My research spans three areas: safety and social dynamics in online platforms, adversarial robustness in federated learning systems, and LLM reasoning and verification. I aim to design AI that is responsible by design and contributes to a safer digital world.
I work with Dr. Yasin Silva and Dr. Mohammed Abuhammad on AI safety and cybersecurity challenges.
Education
Loyola University Chicago
- MS in Data Science
University of the People
- BS in Computer Science
Microverse
- Software Development Program
Recent News (view all )
Teaching
Loyola University Chicago
Teaching Assistant
Mentorship (view all )
Loyola University Chicago
Department of Computer Science
- Ayaan Khan – Undergraduate, Computer Science
- Khushboo Bhadauria – MS, Computer Science
- Josue Torres Romero (Spring 2025) – Undergraduate, Computer Science · Minorities in Tech
Selected Service (view all )
Loyola University Chicago
Graduate Student Advisory Council (GSAC) · CS Department
- Graduate Program Representative, PhD CS (2025 – Present)
- Graduate Program Representative, MS Data Science (Aug 2023 – May 2025)
Chicago Area Undergraduate Research Symposium (CAURS)
- Judge (2025 & 2026)
ICWSM
International AAAI Conference on Web and Social Media
- Reviewer (2025)
SBP-BRiMS
Social Computing, Behavioral-Cultural Modeling & Prediction
- Reviewer (2024 & 2025)
Selected Publications (view all )
SpectrumNet: Detecting LGBTQ+ Cyberbullying with Dynamic Context-Aware Attention
Muhammad Arslan, Manuel Sandoval, Mujtaba Nazari, Mohammed Abuhamad, Deborah L. Hall, Yasin N. Silva
The 20th International AAAI Conference on Web and Social Media (ICWSM)
Cyberbullying remains a critical societal issue, with LGBTQ+ individuals disproportionately affected. Although previous work proposed general cyberbullying detection models, LGBTQ+-targeted cyberbullying detection remains relatively unexplored. SpectrumNet, a novel transformer-based model introduced in this paper, goes beyond conventional cyberbullying detection by adding conversational context and identity-aware modeling. SpectrumNet freezes the RoBERTa backbone and adds three key components: a hierarchical attention network to capture linguistic nuance, a GRU-based encoder to better capture comment history, and a dynamic fusion module to effectively weigh contextual signals. To address dataset imbalance, we apply focal loss and weighted sampling. Trained on a large, annotated Instagram dataset, SpectrumNet effectively differentiates between non-bullying, general bullying, and LGBTQ+-targeted bullying. In particular, it achieves strong recall on targeted content and excels at detecting subtle forms of discrimination often missed in isolation but evident within threaded interactions.
@inproceedings{arslan2026spectrumnet,
title={SpectrumNet: Detecting LGBTQ+ Cyberbullying with Dynamic Context-Aware Attention},
author={Arslan, Muhammad and Sandoval, Manuel and Nazari, Mujtaba and Abuhamad, Mohammed and Hall, Deborah L and Silva, Yasin N},
booktitle={The 20th International AAAI Conference on Web and Social Media},
pages={12},
year={2026}
}
SpectrumNet: Detecting LGBTQ+ Cyberbullying with Dynamic Context-Aware Attention
Muhammad Arslan, Manuel Sandoval, Mujtaba Nazari, Mohammed Abuhamad, Deborah L. Hall, Yasin N. Silva
The 20th International AAAI Conference on Web and Social Media (ICWSM)
Cyberbullying remains a critical societal issue, with LGBTQ+ individuals disproportionately affected. Although previous work proposed general cyberbullying detection models, LGBTQ+-targeted cyberbullying detection remains relatively unexplored. SpectrumNet, a novel transformer-based model introduced in this paper, goes beyond conventional cyberbullying detection by adding conversational context and identity-aware modeling. SpectrumNet freezes the RoBERTa backbone and adds three key components: a hierarchical attention network to capture linguistic nuance, a GRU-based encoder to better capture comment history, and a dynamic fusion module to effectively weigh contextual signals. To address dataset imbalance, we apply focal loss and weighted sampling. Trained on a large, annotated Instagram dataset, SpectrumNet effectively differentiates between non-bullying, general bullying, and LGBTQ+-targeted bullying. In particular, it achieves strong recall on targeted content and excels at detecting subtle forms of discrimination often missed in isolation but evident within threaded interactions.
@inproceedings{arslan2026spectrumnet,
title={SpectrumNet: Detecting LGBTQ+ Cyberbullying with Dynamic Context-Aware Attention},
author={Arslan, Muhammad and Sandoval, Manuel and Nazari, Mujtaba and Abuhamad, Mohammed and Hall, Deborah L and Silva, Yasin N},
booktitle={The 20th International AAAI Conference on Web and Social Media},
pages={12},
year={2026}
}
From Moderation to Mediation: Can LLMs Serve as Mediators in Online Flame Wars?
Dawei Li, Abdullah Alnaibari, Arslan Bisharat, Manny Sandoval, Deborah Hall, Yasin Silva, Huan Liu
PAKDD 2026 Special Session on Data Science: Foundations and Applications
The rapid advancement of large language models (LLMs) has opened new possibilities for AI for good applications. As LLMs increasingly mediate online communication, their potential to foster empathy and constructive dialogue becomes an important frontier for responsible AI research. This work explores whether LLMs can serve not only as moderators that detect harmful content, but as mediators capable of understanding and de-escalating online conflicts. Our framework decomposes mediation into two subtasks: judgment, where an LLM evaluates the fairness and emotional dynamics of a conversation, and steering, where it generates empathetic, de-escalatory messages to guide participants toward resolution. To assess mediation quality, we construct a large Reddit-based dataset and propose a multi-stage evaluation pipeline combining principle-based scoring, user simulation, and human comparison. Experiments show that API-based models outperform open-source counterparts in both reasoning and intervention alignment when doing mediation. Our findings highlight both the promise and limitations of current LLMs as emerging agents for online social mediation.
@inproceedings{li2026moderation,
title={From Moderation to Mediation: Can LLMs Serve as Mediators in Online Flame Wars?},
author={Li, Dawei and Alnaibari, Abdullah and Bisharat, Arslan and Sandoval, Manny and Hall, Deborah and Silva, Yasin and Liu, Huan},
booktitle={PAKDD 2026 Special Session on Data Science: Foundations and Applications},
year={2026}
}
From Moderation to Mediation: Can LLMs Serve as Mediators in Online Flame Wars?
Dawei Li, Abdullah Alnaibari, Arslan Bisharat, Manny Sandoval, Deborah Hall, Yasin Silva, Huan Liu
PAKDD 2026 Special Session on Data Science: Foundations and Applications
The rapid advancement of large language models (LLMs) has opened new possibilities for AI for good applications. As LLMs increasingly mediate online communication, their potential to foster empathy and constructive dialogue becomes an important frontier for responsible AI research. This work explores whether LLMs can serve not only as moderators that detect harmful content, but as mediators capable of understanding and de-escalating online conflicts. Our framework decomposes mediation into two subtasks: judgment, where an LLM evaluates the fairness and emotional dynamics of a conversation, and steering, where it generates empathetic, de-escalatory messages to guide participants toward resolution. To assess mediation quality, we construct a large Reddit-based dataset and propose a multi-stage evaluation pipeline combining principle-based scoring, user simulation, and human comparison. Experiments show that API-based models outperform open-source counterparts in both reasoning and intervention alignment when doing mediation. Our findings highlight both the promise and limitations of current LLMs as emerging agents for online social mediation.
@inproceedings{li2026moderation,
title={From Moderation to Mediation: Can LLMs Serve as Mediators in Online Flame Wars?},
author={Li, Dawei and Alnaibari, Abdullah and Bisharat, Arslan and Sandoval, Manny and Hall, Deborah and Silva, Yasin and Liu, Huan},
booktitle={PAKDD 2026 Special Session on Data Science: Foundations and Applications},
year={2026}
}
ML Model to Better Identify Instances of Bullying Faced by Members Of the LGBTQ+ Community
Loyola University Chicago (Master's Thesis)
Cyberbullying poses a significant threat to online communities, with the LGBTQ+ community facing disproportionately higher rates of harassment. While existing cyberbullying detection systems have made progress in identifying general instances of online harassment, they often fail to capture the nuanced and context-dependent nature of LGBTQ+-targeted bullying. This thesis presents a novel approach to this challenge by developing SpectrumNet, an LGBTQ+-centric transformer-based model for cyberbullying detection. Our research was conducted in two phases. In Phase 1, we evaluated the effectiveness of pre-trained transformer models (RoBERTa, BERT, and GPT-2) in identifying LGBTQ+-related cyberbullying. Building on these findings, Phase 2 introduced SpectrumNet which integrates dynamic attention mechanisms with hierarchical attention networks to understand the contextual nuances of LGBTQ+ targeted harassment.
@mastersthesis{arslan2025ml,
title={ML Model to Better Identify Instances of Bullying Faced by Members Of the LGBTQ+ Community},
author={Arslan, Muhammad},
year={2025},
school={Loyola University Chicago}
}
ML Model to Better Identify Instances of Bullying Faced by Members Of the LGBTQ+ Community
Loyola University Chicago (Master's Thesis)
Cyberbullying poses a significant threat to online communities, with the LGBTQ+ community facing disproportionately higher rates of harassment. While existing cyberbullying detection systems have made progress in identifying general instances of online harassment, they often fail to capture the nuanced and context-dependent nature of LGBTQ+-targeted bullying. This thesis presents a novel approach to this challenge by developing SpectrumNet, an LGBTQ+-centric transformer-based model for cyberbullying detection. Our research was conducted in two phases. In Phase 1, we evaluated the effectiveness of pre-trained transformer models (RoBERTa, BERT, and GPT-2) in identifying LGBTQ+-related cyberbullying. Building on these findings, Phase 2 introduced SpectrumNet which integrates dynamic attention mechanisms with hierarchical attention networks to understand the contextual nuances of LGBTQ+ targeted harassment.
@mastersthesis{arslan2025ml,
title={ML Model to Better Identify Instances of Bullying Faced by Members Of the LGBTQ+ Community},
author={Arslan, Muhammad},
year={2025},
school={Loyola University Chicago}
}
Detecting LGBTQ+ Instances of Cyberbullying
Muhammad Arslan, Manuel Sandoval Madrigal, Mohammed Abuhamad, Deborah Hall, Yasin Silva
17th International Conference on Social Computing, Behavioral-Cultural Modeling & Prediction (SBP-BRIMS)
Social media continues to have an impact on the trajectory of humanity. However, its introduction has also weaponized keyboards, allowing the abusive language normally reserved for in-person bullying to jump onto the screen, i.e., cyberbullying. Cyberbullying poses a significant threat to adolescents globally, affecting the mental health and well-being of many. A group that is particularly at risk is the LGBTQ+ community, as researchers have uncovered a strong correlation between identifying as LGBTQ+ and suffering from greater online harassment. Therefore, it is critical to develop machine learning models that can accurately discern cyberbullying incidents as they happen to LGBTQ+ members. The aim of this study is to compare the efficacy of several transformer models in identifying cyberbullying targeting LGBTQ+ individuals. We seek to determine the relative merits and demerits of these existing methods in addressing complex and subtle kinds of cyberbullying by assessing their effectiveness with real social media data.
@inproceedings{arslan2024detecting,
title={Detecting LGBTQ+ Instances of Cyberbullying},
author={Arslan, Muhammad and Madrigal, Manuel Sandoval and Abuhamad, Mohammed and Hall, Deborah L and Silva, Yasin},
booktitle={17th International Conference on Social Computing, Behavioral-Cultural Modeling, \& Prediction and Behavior Representation in Modeling and Simulation},
year={2024}
}
Detecting LGBTQ+ Instances of Cyberbullying
Muhammad Arslan, Manuel Sandoval Madrigal, Mohammed Abuhamad, Deborah Hall, Yasin Silva
17th International Conference on Social Computing, Behavioral-Cultural Modeling & Prediction (SBP-BRIMS)
Social media continues to have an impact on the trajectory of humanity. However, its introduction has also weaponized keyboards, allowing the abusive language normally reserved for in-person bullying to jump onto the screen, i.e., cyberbullying. Cyberbullying poses a significant threat to adolescents globally, affecting the mental health and well-being of many. A group that is particularly at risk is the LGBTQ+ community, as researchers have uncovered a strong correlation between identifying as LGBTQ+ and suffering from greater online harassment. Therefore, it is critical to develop machine learning models that can accurately discern cyberbullying incidents as they happen to LGBTQ+ members. The aim of this study is to compare the efficacy of several transformer models in identifying cyberbullying targeting LGBTQ+ individuals. We seek to determine the relative merits and demerits of these existing methods in addressing complex and subtle kinds of cyberbullying by assessing their effectiveness with real social media data.
@inproceedings{arslan2024detecting,
title={Detecting LGBTQ+ Instances of Cyberbullying},
author={Arslan, Muhammad and Madrigal, Manuel Sandoval and Abuhamad, Mohammed and Hall, Deborah L and Silva, Yasin},
booktitle={17th International Conference on Social Computing, Behavioral-Cultural Modeling, \& Prediction and Behavior Representation in Modeling and Simulation},
year={2024}
}
1D-CNN-IDS: 1D CNN-based Intrusion Detection System for IIoT
Muhammad Arslan, Muhammad Mubeen, Muhammad Bilal, Saadullah Farooq Abbasi
29th International Conference on Automation and Computing
The demand for Internet of Things (IoT) has seen a rapid increase. These advances have been made possible by technological advances in artificial intelligence, cloud computing, and edge computing. However, these developments present a number of challenges, including cyber threats, security and privacy concerns, and the risk of potential financial losses. For this reason, this study developed a computationally inexpensive one-dimensional convolutional neural network (1DCNN) algorithm for cyber-attack classification. The proposed study achieved an accuracy of 99.90% to classify nine cyber-attacks. Several other performance metrics are evaluated to validate the effectiveness of the proposed scheme. In addition, a comparison has been made with the existing state-of-the-art schemes. The findings of the proposed study can significantly contribute to the development of secure intrusion detection for IIoT systems.
@inproceedings{arslan2024cnn,
title={1D-CNN-IDS: 1D CNN-based intrusion detection system for IIoT},
author={Arslan, Muhammad and Mubeen, Muhammad and Bilal, Muhammad and Abbasi, Saadullah Farooq},
booktitle={2024 29th International Conference on Automation and Computing},
year={2024}
}
1D-CNN-IDS: 1D CNN-based Intrusion Detection System for IIoT
Muhammad Arslan, Muhammad Mubeen, Muhammad Bilal, Saadullah Farooq Abbasi
29th International Conference on Automation and Computing
The demand for Internet of Things (IoT) has seen a rapid increase. These advances have been made possible by technological advances in artificial intelligence, cloud computing, and edge computing. However, these developments present a number of challenges, including cyber threats, security and privacy concerns, and the risk of potential financial losses. For this reason, this study developed a computationally inexpensive one-dimensional convolutional neural network (1DCNN) algorithm for cyber-attack classification. The proposed study achieved an accuracy of 99.90% to classify nine cyber-attacks. Several other performance metrics are evaluated to validate the effectiveness of the proposed scheme. In addition, a comparison has been made with the existing state-of-the-art schemes. The findings of the proposed study can significantly contribute to the development of secure intrusion detection for IIoT systems.
@inproceedings{arslan2024cnn,
title={1D-CNN-IDS: 1D CNN-based intrusion detection system for IIoT},
author={Arslan, Muhammad and Mubeen, Muhammad and Bilal, Muhammad and Abbasi, Saadullah Farooq},
booktitle={2024 29th International Conference on Automation and Computing},
year={2024}
}
A Deep Features Based Approach Using Modified ResNet50 and Gradient Boosting for Visual Sentiments Classification
Muhammad Arslan, Muhammad Mubeen, Arslan Akram, Saadullah Farooq Abbasi, Muhammad Salman Ali, Muhammad Usman Tariq
IEEE 7th International Conference on Multimedia Information Processing and Retrieval
The versatile nature of Visual Sentiment Analysis (VSA) is one reason for its rising profile. It isn’t easy to efficiently manage social media data with visual information since previous research has concentrated on Sentiment Analysis (SA) of single modalities, like textual. In addition, most visual sentiment studies need to adequately classify sentiment because they are mainly focused on simply merging modal attributes without investigating their intricate relationships. This prompted the suggestion of developing a fusion of deep learning and machine learning algorithms. In this research, a deep feature-based method for multiclass classification has been used to extract deep features from modified ResNet50. Furthermore, gradient boosting algorithm has been used to classify photos containing emotional content. The approach is thoroughly evaluated on two benchmarked datasets, CrowdFlower and GAPED.
@inproceedings{arslan2024deep,
title={A deep features based approach using modified ResNet50 and gradient boosting for visual sentiments classification},
author={Arslan, Muhammad and Mubeen, Muhammad and Akram, Arslan and Abbasi, Saadullah Farooq and Ali, Muhammad Salman and Tariq, Muhammad Usman},
booktitle={2024 IEEE 7th International Conference on Multimedia Information Processing and Retrieval},
pages={239--242},
year={2024}
}
A Deep Features Based Approach Using Modified ResNet50 and Gradient Boosting for Visual Sentiments Classification
Muhammad Arslan, Muhammad Mubeen, Arslan Akram, Saadullah Farooq Abbasi, Muhammad Salman Ali, Muhammad Usman Tariq
IEEE 7th International Conference on Multimedia Information Processing and Retrieval
The versatile nature of Visual Sentiment Analysis (VSA) is one reason for its rising profile. It isn’t easy to efficiently manage social media data with visual information since previous research has concentrated on Sentiment Analysis (SA) of single modalities, like textual. In addition, most visual sentiment studies need to adequately classify sentiment because they are mainly focused on simply merging modal attributes without investigating their intricate relationships. This prompted the suggestion of developing a fusion of deep learning and machine learning algorithms. In this research, a deep feature-based method for multiclass classification has been used to extract deep features from modified ResNet50. Furthermore, gradient boosting algorithm has been used to classify photos containing emotional content. The approach is thoroughly evaluated on two benchmarked datasets, CrowdFlower and GAPED.
@inproceedings{arslan2024deep,
title={A deep features based approach using modified ResNet50 and gradient boosting for visual sentiments classification},
author={Arslan, Muhammad and Mubeen, Muhammad and Akram, Arslan and Abbasi, Saadullah Farooq and Ali, Muhammad Salman and Tariq, Muhammad Usman},
booktitle={2024 IEEE 7th International Conference on Multimedia Information Processing and Retrieval},
pages={239--242},
year={2024}
}
A Single Channel-Based Neonatal Sleep-Wake Classification using Hjorth Parameters and Improved Gradient Boosting
Muhammad Arslan, Muhammad Mubeen, Saadullah Farooq Abbasi, Muhammad Shahbaz Khan, Wadii Boulila, Jawad Ahmad
International Polydisciplinary Conference on Artificial Intelligence and New Technologies
Sleep plays a crucial role in neonatal development. Monitoring the sleep patterns in neonates in a Neonatal Intensive Care Unit (NICU) is imperative for understanding the maturation process. While polysomnography (PSG) is considered the best practice for sleep classification, its expense and reliance on human annotation pose challenges. Existing research often relies on multichannel EEG signals; however, concerns arise regarding the vulnerability of neonates and the potential impact on their sleep quality. This paper introduces a novel approach to neonatal sleep stage classification using a single-channel gradient boosting algorithm with Hjorth features. The gradient boosting parameters are fine-tuned using random search cross-validation (randomsearchCV), achieving an accuracy of 82.35% for neonatal sleep-wake classification. Validation is conducted through 5-fold cross-validation. The proposed algorithm not only enhances existing neonatal sleep algorithms but also opens avenues for broader applications.
@inproceedings{arslan2024single,
title={A Single Channel-Based Neonatal Sleep-Wake Classification using Hjorth Parameters and Improved Gradient Boosting},
author={Arslan, Muhammad and Mubeen, Muhammad and Abbasi, Saadullah Farooq and Khan, Muhammad Shahbaz and Boulila, Wadii and Ahmad, Jawad},
booktitle={International Polydisciplinary Conference on Artificial Intelligence and New Technologies},
year={2024}
}
A Single Channel-Based Neonatal Sleep-Wake Classification using Hjorth Parameters and Improved Gradient Boosting
Muhammad Arslan, Muhammad Mubeen, Saadullah Farooq Abbasi, Muhammad Shahbaz Khan, Wadii Boulila, Jawad Ahmad
International Polydisciplinary Conference on Artificial Intelligence and New Technologies
Sleep plays a crucial role in neonatal development. Monitoring the sleep patterns in neonates in a Neonatal Intensive Care Unit (NICU) is imperative for understanding the maturation process. While polysomnography (PSG) is considered the best practice for sleep classification, its expense and reliance on human annotation pose challenges. Existing research often relies on multichannel EEG signals; however, concerns arise regarding the vulnerability of neonates and the potential impact on their sleep quality. This paper introduces a novel approach to neonatal sleep stage classification using a single-channel gradient boosting algorithm with Hjorth features. The gradient boosting parameters are fine-tuned using random search cross-validation (randomsearchCV), achieving an accuracy of 82.35% for neonatal sleep-wake classification. Validation is conducted through 5-fold cross-validation. The proposed algorithm not only enhances existing neonatal sleep algorithms but also opens avenues for broader applications.
@inproceedings{arslan2024single,
title={A Single Channel-Based Neonatal Sleep-Wake Classification using Hjorth Parameters and Improved Gradient Boosting},
author={Arslan, Muhammad and Mubeen, Muhammad and Abbasi, Saadullah Farooq and Khan, Muhammad Shahbaz and Boulila, Wadii and Ahmad, Jawad},
booktitle={International Polydisciplinary Conference on Artificial Intelligence and New Technologies},
year={2024}
}