Dr Orvila Sarker

Research Areas and Interest

Orvila Sarker mainly works in the intersection of Human-Computer Interaction, Usable Security, and Empirical Software Engineering . She published papers in leading software engineering and security venues such as the Journal of Systems and Software, ESEM, and Trustcom conferences. Orvila is passionate about incorporating diversity and inclusion in computing education, software, tools, and technologies. She is interested in developing techniques and solutions using natural language processing and AI to address the usability issues of existing educational and cybersecurity tools and technologies.

Through her PhD, Orvila gained experience both in qualitative and quantitative research methods. She has gathered knowledge in conducting systematic literature reviews, applying qualitative and quantitative research methods such as thematic analysis and grounded theory, conducting semi-structured interviews, and applying Reinforcement Learning/Natural Language Processing techniques in solving cyber security issues and designing user-friendly anti-phishing software and tools.

Research Projects

While working at The University of Adelaide as a Research Associate at CREST, she worked on two projects. The first project focused on developing a Natural Language Processing (NLP) tool prototype aimed at aiding cybersecurity practitioners in designing more effective anti-phishing tools and interventions. This project involved cleaning and summarizing literature data using various prompt engineering techniques and Large Language Models (LLMs), followed by training different NLP models with the curated dataset to achieve the desired outcomes. The second project identified 15 factors that influence the employment success of individuals with Autism Spectrum Disorder (ASD) within the software industry, specifically in the realms of software engineering education, career training programs, and programming tools and assistive technologies.

As a part of her PhD, to assist the developers and practitioners in better designing, implementing, and evaluating the anti-phishing interventions, her research identified 22 human-centric issues in 14 different types of anti-phishing interventions and devised 41 personalized guidelines to mitigate these issues. This work has been published at the 17th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM) 2023 (CORE rank A). Additionally, she has conducted 18 semi-structured interviews with anti-phishing software developers and cyber security practitioners. This study reports anti-phishing practices in organizations, 8 challenges, and requirements in the anti-phishing interventions, extracts features for an envisioned tool, and provides 15 recommendations to improve the current anti-phishing defense in organizations. This interview study has been recently accepted by the Journal of Systems and Software (CORE rank A) with minor revisions.

At the beginning of her PhD, she worked on designing solutions for establishing secure communications among the network nodes in the Vehicular Ad-hoc Network (VANET). This study has been published at the 22nd IEEE International Conference on Trust, Security, and Privacy in Computing and Communications (TrustCom 2023) (CORE rank B). This work proposed a Reinforcement Learning algorithm to prevent four types of cyber attacks.