FGFabian Gilson

Fabian Gilson

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Fabian Gilson has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Identifying Quality Indicators in Student Self-Reflections in Software Engineering

    by Matthew Minish, Matthias Galster, Fabian Gilson

    Context: Reflection is a fundamental skill in software engineering education, particularly in project-based courses where students learn through extended group work and need to develop their ability to reflect iteratively throughout their work. For students to benefit from reflection, their written reflections need to be assessed so that feedback can guide and improve their reflective practice. However, manually assessing written reflections to guide reflections is time-consuming. Furthermore, assessment of reflections often results in broad, non-specific feedback for a student to improve. Objective: This study builds on reflective writing frameworks to produce an eight-indicator scheme for assessing student reflections in software engineering. Furthermore, this study validates an automated classifier for assessing reflections against the framework, enabling scalable and structured feedback whilst reducing instructor workload. Method: We adapted existing reflection frameworks through iterative refinement to create our eight-indicator framework. Three annotators labelled student reflection texts, establishing moderate to reliable inter-rater agreement. We then trained and evaluated multiple encoder-only transformer models and compared them with decoder-only large language models using zero-shot prompting. Results: The fine-tuned RoBERTa model achieved the strongest performance, substantially outperforming decoder-only models in both accuracy and speed. The classifier demonstrated human-level agreement on most indicators whilst enabling near-instantaneous classification. We provide two model variants optimised for different assessment priorities. Conclusions: Our fine-tuned encoder-only models enable efficient automated assessment of reflective writing. The framework and automated classifier offer a means to provide timely, structured feedback on student reflections in software engineering.

  • Open Access

    Improving Software Engineering Team Communication Through Stronger Social Networks

    by April Clarke, Tanja Mitrović, Fabian Gilson

    Students working in teams in software engineering group project often communicate ineffectively, which reduces the quality of deliverables, and is therefore detrimental for project success. An important step towards addressing areas of improvement is identifying which changes to communication will improve team performance the most. We applied two different communication analysis techniques, triad census and socio-technical congruence, to data gathered from a two-semester software engineering group project. Triad census uses the presence of edges between groups of three nodes as a measure of network structure, while socio-technical congruence compares the fit of a team's communication to their technical dependencies. Our findings suggest that each team's triad census for a given sprint is promising as a predictor of the percentage of story points they pass, which is closely linked to project success. Meanwhile, socio-technical congruence is inadequate as the sole metric for predicting project success in this context. We discuss these findings, and their potential applications improve communication in a software engineering group project.

  • Open Access

    Towards Improving Behavior-Driven Development and Acceptance Testing-Driven Development Teaching in a University Project Course

    by Marina Filipovic, Fabian Gilson

    Behavior-Driven Development (BDD) improves the collaboration between developers and stakeholders following agile software development practices. Acceptance Test-Driven Development (ATDD) is an extension of BDD where requirements are accompanied by automated acceptance tests, translating functional acceptance criteria linked to requirements. While BDD and ATDD have been successfully applied in industry, few universities report on teaching BDD and ATDD principles as part of their curriculum. However, there is a need for education providers to teach industry-relevant practices so that junior developers’ skills are better aligned to industry’s expectations. Therefore, we report on how BDD and ATDD techniques have been taught in a two-semester software engineering project, including (i) the lack of observed evidence on the implications of applying BDD and ATDD on the functional suitability of software products, (ii) the lack of engagement with BDD and ATDD from students, and (ii) recommendations to educators to improve students’ engagement with these practices.

  • Open Access

    Which Combination of Test Metrics Can Predict Success of a Software Project? A Case Study in a Year-Long Project Course

    by Marina Filipovic, Fabian Gilson

    Testing plays an important role in securing the success of a software development project. Prior studies have demonstrated beneficial effects of applying acceptance testing within a Behavioural-Driven Development method. In this research, we investigate whether we can quantify the effects various types of testing have on functional suitability, i.e. the software conformance to users' functional expectations. We explore which combination of software testing (automated and manual, including acceptance testing) should be applied to ensure the expected functional requirements are met, as well as whether the lack of testing during a development iteration causes a significant increase of effort spent fixing the project later on. To answer those questions, we collected and analysed data from a year-long software engineering project course. We combined manual observations and statistical methods, namely Linear Mixed-Effects Modelling, to evaluate the effects of coverage metrics as well as time effort on passed stories over 5 Scrum sprints. The results suggest that a combination of a high code coverage for all of automated unit, acceptance, and manual testing has a significant impact on functional suitability. Similarly, but to a lower extent, front-end unit testing and manual testing can predict the success of a software delivery when taken independently. We observed a close-to-significant effect between low back-end testing and deferral (i.e. postponement) of user stories.

  • Open Access

    A Three-Year Study on Peer Evaluation in a Software Engineering Project Course

    by Miguel Ehecatl Morales-Trujillo, Matthias Galster, Fabian Gilson, Moffat Mathews

    Background: Peer evaluation in software engineering (SE) project courses enhances the learning experience of students. It also helps instructors monitor and assess both teams and individual students. Peer evaluations might influence the way individual students and teams work; therefore, the quality of the peer evaluations should be tracked through the project course. Contribution: In this article, we analyzed the quality and scoring behavior of students in peer evaluation in an undergraduate SE project course over three years. Research Questions: RQ1: What is the quality of peer evaluation of undergraduate students in a SE project course? RQ2: How do undergraduate students in an SE project course score each other? Methodology: The quality of peer evaluation (length, level of detail, etc.) and scoring of peers based on various aspects of peer evaluations of third-year students in a year-long SE project course were studied. Taking into account the grade students received at the end of the course (A, B, C, and F-calibers), peer evaluations were categorized, analyzed over time, and compared between students calibers. Findings: After analyzing 6854 peer evaluations from 193 students, it was found that the quality of peer evaluations across students was mostly consistent throughout the course. Also, it was observed that quantitative aspects of the peer evaluation were scored similarly across student calibers. However, the qualitative aspects of the peer evaluation were impacted by the caliber of students. These findings suggest that weaker students (i.e., C-caliber students) generally receive better quality peer evaluations than stronger students (i.e., A-caliber students). Finally, a preliminary analysis showed a positive connotation of emotions and sentiments found in the textual feedback delivered by students.


  • Open Access

    Topic modeling in software engineering research

    by Camila Costa Silva, Matthias Galster, Fabian Gilson

    Topic modeling using models such as Latent Dirichlet Allocation (LDA) is a text mining technique to extract human-readable semantic “topics” (i.e., word clusters) from a corpus of textual documents. In software engineering, topic modeling has been used to analyze textual data in empirical studies (e.g., to find out what developers talk about online), but also to build new techniques to support software engineering tasks (e.g., to support source code comprehension). Topic modeling needs to be applied carefully (e.g., depending on the type of textual data analyzed and modeling parameters). Our study aims at describing how topic modeling has been applied in software engineering research with a focus on four aspects: (1) which topic models and modeling techniques have been applied, (2) which textual inputs have been used for topic modeling, (3) how textual data was “prepared” (i.e., pre-processed) for topic modeling, and (4) how generated topics (i.e., word clusters) were named to give them a human-understandable meaning. We analyzed topic modeling as applied in 111 papers from ten highly-ranked software engineering venues (five journals and five conferences) published between 2009 and 2020. We found that (1) LDA and LDA-based techniques are the most frequent topic modeling techniques, (2) developer communication and bug reports have been modelled most, (3) data pre-processing and modeling parameters vary quite a bit and are often vaguely reported, and (4) manual topic naming (such as deducting names based on frequent words in a topic) is common.


  • Open Access

    Topic modeling in software engineering research

    by Camila Costa Silva, Matthias Galster, Fabian Gilson