Computer Science Research Projects 2023
Zhuo (Cecilia) Chen
Advisor: Chris Murphy
Debugging Visualization
Locating bugs in software development is a time-consuming and costly process that involves extensive testing, including failing test cases. In the study "Visualization of Test Information to Assist Fault Localization," Jones et al. proposed a method that displays lines of code in different colors based on the likelihood of containing bugs. In this research, we aim to improve upon this approach by introducing two key modifications. Firstly, we propose displaying the code as a Control Flow Graph, providing users with a clear visualization of the code's path likely to contain bugs. Secondly, we suggest using varying colors and sizes for elements in the Control Flow Graph to indicate the likelihood of containing bugs. Drawing upon insights from information visualization and cognitive psychology, our objective is to present information visually, enabling developers to swiftly identify and resolve bugs. To evaluate the effectiveness of our approach, we will conduct two experiments: a user study to assess the efficiency and accuracy of fault identification and an experiment to determine the accuracy of code highlighting for bug localization. The outcomes of this research will contribute to advancing bug visualization techniques and enhancing the efficiency of the debugging process in software development projects.
Joseph Kim
Advisor: Chris Murphy
Experiences of students with different deadline policies in CS classrooms
During the coronavirus pandemic, the dynamics of classrooms underwent drastic changes. Traditional face-to-face class meetings transitioned to remote learning, requiring both instructors and students to adapt to this sudden shift while managing their personal circumstances. Recognizing the challenges faced by students, instructors granted increased flexibility to help them manage their workload. However, as the pandemic recedes and students return to in-person classes, a lingering question persists in both the classroom and institutional settings: to what extent should students be given flexibility?
On one hand, much research demonstrates that providing students with extensions or allowing for late submissions can enhance learning and academic performance, while simultaneously reducing student stress. This approach grants students a greater sense of control and offers better support to nontraditional students in community colleges and online learning platforms who may have extenuating circumstances. On the other hand, a lack of structure can be detrimental to students prone to procrastination, and many students rely on deadlines to stay on track with their coursework. The impact of a deadline policy is particularly significant in the field of Computer Science, where regular assignment submissions are crucial for assessing student learning and providing necessary feedback.
Hence, the objective of this research is to explore the advantages and disadvantages of various deadline policies in Computer Science classes, with a focus on identifying the policy that yields the best outcome for students. This not only includes academic performance but also learning, time management, and mental health of the students. The findings of this study will enable Computer Science instructors to make more informed decisions regarding the structure of their courses, ultimately benefiting their students. To achieve this goal, a survey will be designed to gather student feedback on their experiences with different deadline policies across multiple dimensions. Additionally, an optional comment section will provide participants with an opportunity to provide more detailed descriptions of their experiences.