Maryam Abbasalizadeh

Maryam Abbasalizadeh

Auto-generated profile, based on publicly available information

Maryam Abbasalizadeh

Miner School of Computer and Information Sciences, University of Massachusetts Lowell, Lowell, MA, USA

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 October 2024 in IEEE Open Journal of the Communications Society
Verified
Journal Article
Open access
MAMaryam Abbasalizadeh
Maryam Abbasalizadeh
KVKrishnaa Vellamchety
Krishnaa Vellamchety
PRPranathi Rayavaram
Pranathi Rayavaram
SNSashank Narain
Sashank Narain
In this paper, we present the Learning Greedy Link Scheduling (LGLS) algorithm, a novel approach for optimizing link scheduling in wireless networks. By integrating deep learning and fuzzy logic, LGLS predicts link quality probabilities, which provide critical topological information to dynamically manage wireless network interference. This approach enhances resource allocation efficiency, leading to better bandwidth and spectrum usage. Our comprehensive evaluation shows that LGLS outperforms traditional algorithms such as Local Greedy Scheduling (LGS), achieving link scheduling performance improvements ranging from 9.60% to 24.79% and activating up to 24.10% more links. These results demonstrate LGLS’s robustness and efficiency in diverse network conditions, making it a promising solution for future wireless network optimization.

10 July 2025 in IEEE Access
Verified
Journal Article
Open access
MAMaryam Abbasalizadeh
Maryam Abbasalizadeh
SNSashank Narain
Sashank Narain
Large Language Models (LLMs) have transformed natural language processing, enabling applications such as conversational agents and machine translation. However, their deployment introduces significant privacy concerns, including the memorization and unintended disclosure of sensitive data. Existing privacy-preserving techniques—such as Differential Privacy and federated learning—struggle to balance data protection, model utility, and computational efficiency. To address these limitations, we propose a lightweight privacy-disclosure detection system that combines Bidirectional Long Short-Term Memory (BiLSTM) networks with Hidden Markov Models (HMM) using a novel modeling pipeline. Our approach employs the Predefined and Sensitive Labeling (PSL) technique, a generative labeling approach that extracts meaningful patterns from data. These patterns are then used to train a BiLSTM model capable of proactively identifying sensitive information in real-time user interactions with LLMs. As BiLSTMs do not provide the probability of private data, we design a HMM that estimates the probability of occurrence for this private data. Utilizing the Forward algorithm, our system quantifies privacy risks, enabling users to revise inputs prior to submission and thereby enhancing data privacy. Trained on synthetic data using PSL technique, the model achieves approximately 99.94% precision, recall, and F1-score, and successfully detects previously unseen sensitive information in synthetic datasets with ≈99.98% accuracy across 55,000 sentences. Additionally, the generated model trained on patterns derived from synthetic data, achieved ≈99.99% accuracy when evaluated on a real-world dataset across varying sentence structures, demonstrating strong generalizability in detecting sensitive information regardless of the data source. Importantly, the model provides real-time predictions with an average execution time of 35.46 milliseconds, satisfying the speed requirements for practical deployment. It also trains 45.5 times faster than a state-of-the-art framework, offering high computational efficiency without compromising accuracy.

12 February 2025
Verified
Conference Paper
Open access
PRPranathi Rayavaram
Pranathi Rayavaram
OUOnyinyechukwu Ukaegbu
Onyinyechukwu Ukaegbu
MAMaryam Abbasalizadeh
Maryam Abbasalizadeh
KVKrishna Vellamchetty
Krishna Vellamchetty
SNSashank Narain
Sashank Narain
This paper presents an educational tool designed to enhance cryptography education for K-12 students, utilizing Kolb's Experiential Learning (EL) model and engaging visual components. Our tool incorporates the four stages of EL-Concrete Experience, Reflective Observation, Abstract Conceptualization, and Active Experimentation-to teach key cryptographic concepts, including hashing, symmetric cryptography, and asymmetric cryptography. The learning experience is enriched with real-world simulations, customized AI-based conversation agents, video demonstrations, interactive scenarios, and a simplified Python coding terminal focused on cryptography. Targeted at beginners in cybersecurity, the tool encourages independent learning with minimal instructor involvement. An evaluation with 51 middle and high school students showed positive feedback from 93% of participants, who found the simulations, visualizations, AI reflections, scenarios, and coding capabilities engaging and conducive to learning. Comprehension surveys indicated a high understanding of cryptography concepts: hashing (middle school: 89%, high school: 92%), symmetric cryptography (middle school: 93%, high school: 97%), and asymmetric cryptography (middle school: 91%, high school: 94%)

6 August 2024 in IEEE Access
Verified
Journal Article
Open access
MAMaryam Abbasalizadeh
Maryam Abbasalizadeh
JCJeffrey Chan
Jeffrey Chan
PRPranathi Rayavaram
Pranathi Rayavaram
YCYimin Chen
Yimin Chen
SNSashank Narain
Sashank Narain
Wireless communication is now a cornerstone of modern society, propelled by the widespread adoption of IoT devices and sophisticated wireless technologies. As wireless networks grow in complexity, there is an increasing need for efficient scheduling algorithms that can manage resources and adapt to evolving conditions effectively. Link scheduling is critical as it optimizes the use of wireless spectrum and bandwidth, prioritizing high-priority transmissions and minimizing interference. However, traditional link scheduling algorithms have primarily focused on performance, often overlooking the crucial aspect of privacy. This oversight poses significant risks in scenarios where privacy is paramount. To address this issue, we introduce PriLink, a novel link scheduling algorithm that prioritizes privacy without sacrificing performance. PriLink employs a least privilege model, sharing only essential links to protect critical topology details from potential adversaries. Our comprehensive evaluation shows that PriLink not only matches but occasionally surpasses the performance of established benchmarks like Greedy Maximal Scheduling and Local Greedy Scheduling. Moreover, it offers faster execution times and superior privacy protection. These results highlight PriLink’s effectiveness as a robust solution for efficient, real-time, and privacy-preserving link scheduling in dynamic wireless networks.

12 February 2025
Verified
Conference Paper
Open access
SNSashank Narain
Sashank Narain
PRPranathi Rayavaram
Pranathi Rayavaram
CMChristopher Morales-Gonzalez
Christopher Morales-Gonzalez
MHMatthew Harper
Matthew Harper
MAMaryam Abbasalizadeh
Maryam Abbasalizadeh
XFXinwen Fu
Xinwen Fu
The increasing sophistication of cybersecurity threats necessitates an educational approach that blends theoretical knowledge with practical experience. Many courses focus primarily on theoretical concepts, leaving students with limited hands-on experience with real-world challenges. This paper introduces a cybersecurity course model that integrates Experiential Learning Theory to provide a comprehensive hands-on learning environment. The course covers important cybersecurity topics, including SSH, VPNs, TLS, MFA, OpenID Connect, OAuth2, web server security, high availability, replication, distributed file systems, and orchestration with Docker and Kubernetes. These topics are explored through a mix of lectures, peer presentations, and weekly hands-on team practices. Over three years, the course has been offered at our large public university with 72 students enrolled, consistently receiving high course ratings between 4.8 and 5.0. This paper discusses the course design, methodology, and outcomes, offering insights for educators to replicate and adapt the model for their own institutions.

7 March 2024
Verified
Conference Paper
Open access
PRPranathi Rayavaram
Pranathi Rayavaram
SDSahithi Charitha Dindukuri
Sahithi Charitha Dindukuri
KVKrishna Vellamchety
Krishna Vellamchety
JMJustin Marward
Justin Marward
MAMaryam Abbasalizadeh
Maryam Abbasalizadeh
SNSashank Narain
Sashank Narain
Role-playing and visual interfaces are vital in computer science education, notably cybersecurity, where a concrete understanding of defensive and offensive strategies matters. In this work, we implemented a cryptography education tool combining real-world scenarios, role-playing and visual interfaces to simplify cryptography education for K-12 students. This tool covers symmetric and asymmetric cryptography and secure hashing algorithms, which are fundamental to principles like confidentiality, integrity protection, and authentication. Our tool's core idea is to simulate a cryptographic transaction between a user and a bank while an attacker tries to compromise the communication. Within this role-playing and visual environment, students step into the roles of both the user and the bank, using an intuitive visual interface to perform essential cryptographic operations and grasp the underlying concepts. We evaluated our educational tool on 58 students (41 middle and 17 high school students), measuring comprehension via a survey. Our analysis revealed that middle and high school students could readily understand the key operations related to symmetric cryptography, asymmetric cryptography, and hashing using our role-playing tool. Nearly all participants expressed satisfaction with the tool and were willing to recommend it to others.