Supervisors and Areas for Thesis Topics

Francisco Gomes

francisco.gomes@cse.gu.se

Research interests: automated software testing and analysis, automated test selection and prioritisation, data analytics, test optimisation, software development bots (DevBots).

Open thesis topics with company collaboration for VT2021:

With company: Zenseact
- Analysis of mutation testing on development of autonomous software: https://career.zenseact.com/jobs/1404599-master-thesis-evaluation-of-mutation-based-testing

Five examples of recently supervised theses:

Evaluating the trade-offs of diversity-based test prioritization: an experiment (2020, Bachelor Thesis)
An experimental study on combining automated and stochastic test data generation (2020, Master's Thesis)
The effect of test case design in software testing bots (2019, Bachelor thesis published at BotSE 2020)
A Case Study on Test Optimisation and Visualisation of Diversity Information (2019, Master's Thesis with Volvo Cars)
Visualisation of Test Failure Data to Support Fault Localisation in Distributed Embedded Systems within the Automotive Industry (2017, BSc Thesis with Volvo Cars- published at ISSRE 2018)

Gregory Gay

ggay@chalmers.se

Research interests: automated software testing and analysis, search-based software engineering, automated test generation, fuzzing, development/testing for environmental sustainability, data analytics, optimization, information retrieval.

Open thesis topics with company collaboration for VT2022:

(Zenseact, co-supervised with Francisco Gomes) Analysis of Mutation Testing on Development of Autonomous Software
(Ericsson) Fuzz-testing Using Gradient-based Methods

Five examples of recently supervised theses:

Jan-Philipp Steghöfer

jan-philipp.steghofer@gu.se

Research interests: Software and Systems Traceability; Agile Software Development; Change Management; Adaptive Systems; Safety and Security; Product Line Engineering; Trust

Open thesis topics for VT2022:

A query language for traceability models: enable engineers to define custom analysis of existing traceability links between artifacts.
An experimentation environment for trace links recovery: enable researchers to try out different trace recovery algorithms directly in the development environment.
A comparison of traceability datasets: provide an overview of the traceability datasets that exist in the wild and how they compare to each other using a set of metrics.
Metrics for traceability models: define metrics that allow analysing the quality of a traceability model and provide feedback about the traceability links to practitioners.
Lots more topics about traceability and other topics. Email me if you want to know more!

Examples of recently supervised theses:

Introducing traceability in an agile safety critical development environment (with 1928diagnostics)
Creating a pattern for security assurance cases in the medical domain (with 1928diagnostics)
Timing Verification of Application Software in Multi-Core Systems (2019, Master's Thesis with Volvo Trucks)
A study of how trust can be assigned in crowdsourced information retrieval (2018, Master's Thesis)
Agile Practices’ Relationship to Barriers in Global Software Development (2018, Bachelor's Thesis, led to a paper presented at the International Conference on Software Engineering)

Jennifer Horkoff

jenho@chalmers.se

Research interests: requirements engineering; conceptual modeling; software value; creativity; business intelligence; APIs;

Open thesis topics with company collaboration for VT2021:

Possibly some topics with companies, but also more lab-oriented studies
http://www.cse.chalmers.se/~jenho/Teaching/Theses/ThesisTopics2022.html

Examples of recently supervised theses:

Selecting Business Intelligence Solutions using Requirements Engineering (2019, Master's Thesis)
Evaluating and Developing Methods to Assess Business Process Suitability for Robotic Process Automation: A Design Research Approach (2018, Master's Thesis)
Incremental requirements engineering in a large-scale agile and safety-critical context: a case study (2018, Master's Thesis)
A UML Activity Diagram Extension and Template for Bioinformatics Workflows: A Design Science Study (2019, Bacher's Thesis)
A Framework for Applying an API-layered Architecture to e3 value Models (2018, Bachelor's Thesis)
Find the full list here

Miroslaw Staron

miroslaw.staron@gu.se

Research interests: data analytics; software metrics; software design; mining software repositories; programming languages and techniques; automotive software engineering; telecom software engineering; continuous integration;

Open thesis topics with company collaboration for VT2020:

Volvo Cars: Using Machine Learning for triaging of incident data (software maintenance)
Ericsson: Software measurement optimized data storage
Swedco: Use of Artificial Intelligence to improve the digitization process in urban development
Other thesis topics also available with Software center companies - www.software-center.se

Examples of recently supervised theses:

Applying Machine Learning to Identify Maintenance Level for Software Releases (2019, Master's Thesis)
Improving the Continuous Integration Process for the Vehicle Dynamics Software at VCC using Docker (2019, Master's Thesis)

Philipp Leitner

philipp.leitner@chalmers.se

Research interests: Cloud computing; software performance (especially for Web- and cloud-based systems); modern release engineering (canary releases, A/B testing, chaos engineering, etc.); DevOps; mining software repositories

See "Joel Scheuner" and "Linda Erlenhov" for more thesis topics from my group.

Open master's thesis topics:

Optimizing the generation of Java JMH Benchmarks (topic 1 here)
Predicting Optimal Cloud Instances Using Multi-Armed Bandits (the first topic here)
Predicting Performance Variability of Go Microbenchmarks using Graph Neural Networks (the second topic here)

Open bachelor thesis topics:

Qualitative and Quantitative Comparison of Benchmarking Frameworks Across Languages (topic 2 here)

Examples of recently supervised theses:

The Impact of Adopting Continuous Integration on the Delivery Time of Pull Requests - A Partial Replication and Extension (2019, Master's Thesis)
Automatic container profiling and resource prediction using container orchestration (2019, Master's Thesis)
An Extensible Framework for Large-Scale Mining of Performance Microbenchmarks (2018, Master's Thesis)
Cachematic - Automatic Invalidation in Application-Level Caching Systems (2018, Master's Thesis)

Joel Scheuner

scheuner@chalmers.se

joelscheuner.com

Research interests: Cloud computing (especially serverless); performance engineering (especially measuring the performance of cloud systems); cloud DevOps (especially software engineering for Infrastructure-as-Code); mining software repositories (especially for cloud applications)

Open thesis topics with company collaboration for VT2022:

None

Ideas of thesis topics for VT2022:

Topic 1: Serverless Cost Estimation
- Problem: Reduced hosting cost is one of the main drivers for serverless adoption [1] but the cheap per-request costs can go out of control when scaling out [2]. Unfortunately, the total hosting cost of a serverless application is hard to estimate during development.
- Goal: The goal of this thesis is to develop and evaluate a cost estimation technique for serverless applications based on distributed tracing [3] similar to Costradamus [4].
- Getting started: We have a tool for serverless experimentation including sample applications for your evaluation. We also have experience with serverless tracing as well as previous master theses on related topics.
- Requirements: Good programming skills in Node.js, Java, or Python; Motivation to learn new cloud technology; Helpful to know some data science libraries for data analysis and plotting (e.g., Pandas, ggplot)
Topic 2: Workload Classification of Serverless Function Traces ASSIGNED
- Problem: Different workloads (i.e., varying number of users over time) can impact the performance and cost of cloud applications. Classic cloud workloads [1] are well researched but only two recent datasets exist for serverless functions [2, 3].
- Goal: The goal of this thesis is to develop and evaluate a classification technique to automatically identify different classes of workload patterns from real-world invocation traces.
- Getting started: We have references to real-world datasets and a tool to pre-process these traces.
- Requirements: Data analysis and statistics
Topic 3: On the Comparability of Serverless Performance Metrics: compare different existing datasets that report serverless cold start performance based on an existing literature review [1]. Develop guidance on reporting performance datasets.
Topic 4: An Update to the Multi-vocal Literature Review of Function-as-a-Service Performance Evaluation [1]: Following guidance on when to update literature reviews [2]

⇒ Email me if you are motivated to work on any above-listed topic or pitch your own related topic.

UPDATE: I am fully booked for thesis projects this year and cannot supervise further students. If you are interested in related topics, check out the topics listed by Philipp Leiter.

Examples of recently supervised theses:

An In-depth Benchmarking of Real-world Serverless Application Performance Across Cloud Providers to Understand The Reasons for Differences (starting, Master Thesis)
Multi-level FaaS application deployment optimization (ongoing, Master Thesis)
Performance Comparison of Function-as-a-Service Triggers (2021, Master Thesis)
Deployment Failures in Open-Source Serverless Applications (2021, Bachelor Thesis)
Feedback-Driven Function Fusion in FaaS Deployments (2020, Master Thesis)
Serverless Development Trends in Open Source: a Mixed-Research Study (2019, Bachelor Thesis)
AWS Lambda Language Performance (2018, Bachelor Thesis)
Check my website for updates and recent publications: joelscheuner.com

Regina Hebig

regina.hebig@cse.gu.se

hebig@chalmers.se

Research interests: Software Comprehension and Maintenance; Hybrid Software Processes and Metrics; Model-Driven Engineering and DSLs; DevBots for explaining systems

Ideas of thesis topics for VT2022:

Automatically mapping model elements to the modeled code by analyzing Github data
Studying Developer Memory (replicating/extending this study)
CI-based Documentation Generation

Examples of recently supervised theses:

Onboarding in Agile Software Development Teams: A Case Study (2020, Master’s Thesis)
Facilitation of Regular Communication between UI Designers and Developers through a Continuous Pipeline Tool A Design Science Study (2020, Master’s Thesis)
A Case Study of the Challenges with Applying Machine Learning in Industry: A Software Engineering Perspective (2019, Master's Thesis)
Extracting Microservices from a Monolithic Application (2019, Master's Thesis)
Usability-driven development frameworks for DSLs in small-scale projects: A design science study (2019, Bachelor's Thesis)
Deviations From Scrum Framework: Root Causes And Consequences (2019, Master's Thesis)

Richard Berntsson Svensson

richard@cse.gu.se

Research interests: Data Driven Decision Making; Decision Making; Creativity and Innovation; Requirements Engineering; Agile and Lean Software Development.

Open thesis topics with company collaboration for VT2021:

Not for VT21

Examples of recently supervised theses:

Systematic Data-Driven Decision Making in Requirements Engineering (2020, Master's Thesis)
LeAgile Measurement and Metrics: A Systematic Literature Review and Case Study (2017, Master's Thesis)
Gamification as a Tool to Encourage Eco-Driving (2017, Master's Thesis)
Aligning Legacy Requirement Specifications and Semi-Automated Testing (2015, Master's Thesis)
Impact of Usability on Process Lead-Time in Information Systems (2015, Master's Thesis)
More examples can be found on my page

Picture of Richard

Richard Torkar

richard.torkar@cse.gu.se

Research interests: applications of analysis methods in software engineering activities; software testing; soft issues in software engineering; fault prediction; quality prediction; machine learning applications in software engineering; causal analysis; experimental designs

Open thesis topics for VT2022:

Implement sensitivity analysis concerning omitted variable bias in causal analysis (observational studies)
Large scale test case prioritization using Bayesian multilevel models
Stochastic block models (latent networks) for political security data of different measurement types. For an overview please see this. In the end we might rely on including this too: Explaining and predicting security policy changes using text mining methods (this thesis proposal must be adapted to a software engineering context)
The company lynkco.com has intern positions and is interested also in software engineering thesis students. In their own words, it's about the "mobile app platform at Lynk & Co on Lindholmen in Gothenburg where we work on car sharing, digital keys, car connectivity, and much more."
Data analytics with Volvo Cars. Connecting team maturity to productivity measures. Lucas Gren, Martin Shepperd, and yours truly will be very involved in this topic. Martin will be a guest professor with us in 2022.
The company Centiro has intern positions and is interested in students doing a master thesis in ML. In particular, they're looking for a student that will apply Bayesian data analysis to big data in the logistics sector (this thesis proposal must be adapted to a software engineering context)
- Predict production incidents using ML from large log-files - Centiro
- Evaluation of information security policies in relation to standard requirements - Centiro

Examples of recently supervised theses:

The broken windows theory in software engineering: An experiment on technical debt propagation (2021, Master's thesis)
Non-functional requirements testing in the Medtech industry: An automated test framework for vascular surgery simulations (2021, Master's thesis)
Increasing End-Users' Awareness of Software's Environmental Impact: An Exploratory Case Study (2021, Master's thesis)
Measuring affective states from technical debt: A psychoempirical software engineering experiment (2019, Master's thesis) https://arxiv.org/abs/2009.10660
An empirical study of Linespots: A novel past-fault algorithm (2019, Master's thesis) https://arxiv.org/abs/2007.09394

Picture of Prof. R. Feldt

Robert Feldt

robert.feldt@chalmers.se

Research interests: software testing and quality, automated software engineering, human factors & psychology, software teams, machine learning & artificial intelligence, optimisation & search, information theory & statistics & visualisation

Open thesis topics with company collaboration for VT2020:

Optimizing the Diversity and Diameter of Test Sets (with ReTest)
Automated Boundary Testing (with ReTest)
Selecting Typical Test Cases from a Large and Generated Set (with ReTest)
Collective Intelligence of Software Teams (with SAAB and Volvo/VCC)

Examples of recently supervised theses:

David Gustafsson and Erik Pihl, "Strategy for Cross-Project Defect Prediction Models in Industry", in collaboration with CPAC Systens AB. (2018, Master's thesis)
Patrik Haar and David Michaelsson, "Automated GUI Testing: A Comparison Study With A Maintenance Focus", in collaboration with Canea AB. (2018, Master's thesis)
Kevin Hedberg Griggith and Erik Nguyen, "Understanding the Impacts of Affects on Individual and Team Performance with the Experience Sampling Method", in collaboration with Volvo AB. (2018, Master's thesis)
More examples can be found on my MT page

Sergio Garcia

sergio.garcia@gu.se

Research interests: Robotics Software Engineering, Autonomous Systems, Software Architecture, Model-Driven Engineering and DSLs, Variability

Open thesis topics with company collaboration for VT2020:

None

Examples of recently supervised theses:

None

Eric Knauss

eric.knauss@cse.gu.se

oerich.wordpress.com

Research interests: Requirements engineering, architecture, knowledge management, strategic decision making in scaled-agile system development and ai-based system development

Open thesis topics with company collaboration for VT2021:

Prototype and Requirements Engineering of an AI-based Augmented Reality information system for Smart-Build Environments. The prototype should use mobile and augmented reality technology to make AI-based decision support available to key stakeholders on-site. Various partners in the building sector.
Method and approach to describe use cases for autonomous drive and active safety technology in relation to contextual information. In collaboration with a large local supplier of autonomous drive and active safety technology.
Method and approach to describe quality attributes of data in distributed deep learning architectures. In collaboration with a large local supplier of autonomous drive and active safety technology.

Examples of recently supervised theses:

Mebrahtom Guesh Gebremichael (2019): Requirements Engineering for Large-Scale Agile System Development: A Tooling Perspective. Master’s thesis, Dept. of Computer Science and Engineering, Chalmers | University of Gothenburg, Sweden.
Felix Lissåker, Johan Sjöberg (2019): Architecture Framework for Blockchain Implementation: A Design Science Study. Master’s thesis, Dept. of Computer Science and Engineering, Chalmers | University of Gothenburg, Sweden.
Odzaya Batsaikhan, Yi-Chun Lin (2018): Building a Shared Understanding of Customer Value in a Large- Scale Agile Organization. Master’s thesis, Dept. of Computer Science and Engineering, Chalmers | University of Gothenburg, Sweden.
Elsa Mjöl Bergsteinsdóttir, Henrik Helén Edholm (2018): Designing Continuous Toolchains: Using Propo- sed Guidelines and Tool Capabilities. Master’s thesis, Dept. of Computer Science and Engineering, Chalmers | University of Gothenburg, Sweden.
Rim Shahin, Catarina Curta (2018): Designing a Requirements Elicitation Approach for Intelligent and Interactive Systems in Autonomous Vehicles. Master’s thesis, Dept. of Computer Science and Engineering, Chalmers | University of Gothenburg, Sweden.

Alexander Stotsky

alexander.stotsky@chalmers.se

Research interests: Statistical methods, 6-sigma, Data analysis, Numerical methods, Data driven modeling, System identification, Adaptive control, Frequency estimation, Fault detection, Wind energy, Power systems

Open thesis topics for VT2021:

Application of the Statistical Methods in Software Engineering: Systematic Literature Review, Challenges and Future Directions. The topic of the thesis is associated with the literature review, collection of statistical case studies (real-world examples) and determination of new trends in application of the statistical methods in software engineering.

Khaled
Al-Sabbagh

Khaled.al-sabbagh@gu.se

Research interests: Software Testing, Continuous Integration, Machine learning for Software Engineering, Software teams and group maturity.

Open thesis topics with company collaboration for VT2021:

None

Ideas of thesis topics for VT2021:

Reducing The Cost Of Continuous Integration Using a Build Selection Approach.
Improving Code Violation Detection Using Machine Learning and Noise Handling Techniques.
Identifying Dependency Patterns Between Test Case Outcomes and Code Change Categories.

Fadhl Hujainah

fadhl@chalmers.se, fadelhogina@gmail.com

Research interests: Requirements Engineering, Architecture, Decision Making, Stakeholder Analysis

Open thesis topics for VT2022:

Tooling for Conformance checking code vs. architectural design
A technique and tool for software requirements prioritization using AI Algorithm.
A technique and tool for stakeholder quantification and prioritization using AI Algorithm.
Method to automize the code review based on AI technique.
Method for evaluating the code review assessments.
A Systematic Literature Review on Scalability Issue in Software Requirements Prioritization

Christian Berger

christian.berger@gu.se

Research interests: Architecting complex and distributed realtime software systems, micro-services for embedded systems and cyber-physical systems, continuous integration/deployment/experimentation, autonomous driving, simulations, scalability for real-time data streams.

Open thesis topics for 2021/2022:

Designing, training, and applying ML-models to spot interesting traffic situations on a dataset covering more than 1 year of highway driving data covering video streams, vehicle onboard data, and object providing sensors.
Systematically evaluating the effect of lossy video compression algorithms on the performance of state-of-the-art ML models using the Waymo Open Dataset.
In collaboration with Chi Zhang (PhD candidate in SHAPE-IT, www.shape-it.eu, https://www.shape-it.eu/esr/esr-3-classifying-and-predicting-interactions-between-av-and-vrus-using-ai/)
- Based on this paper: https://openaccess.thecvf.com/content_ICCV_2019/papers/Rasouli_PIE_A_Large-Scale_Dataset_and_Models_for_Pedestrian_Intention_Estimation_ICCV_2019_paper.pdf , which is working on the prediction of pedestrian's behavior including trajectory and intention and that has been initially explored in a previous Bachelor thesis project, we aim at investigating further the pedestrian's intention.
  - What has been accomplished in the Bachelor thesis project:
    - replicating the PIE algorithm and LSTM network on PIE Dataset
    - exploring the possibility of using it on the Waymo Open Dataset
    - adding manual labels
  - How to continue from here:
    - Replicate the state-of-the-art intention prediction model. A benchmark paper proposed this year mentioned several models, we can choose one or several of it: https://openaccess.thecvf.com/content/WACV2021/papers/Kotseruba_Benchmark_for_Evaluating_Pedestrian_Action_Prediction_WACV_2021_paper.pdf
    - Explore the datasets that can be used for the intention prediction. Currently, widely used datasets include: JAAD, PIE. During the Bachelor thesis project, labeling has been conducted for the Waymo Open Dataset and this data can be reused.
    - Propose / Design the model/network that can improve existing intention/behavior prediction, by, for example, using intention to improve trajectory predictions, or conversely; learn the interactions between different traffic agent, etc.

Birgit Penzenstadler

birgitp@chalmers.se

Research interests: sustainability; neuroplasticity practices for resilience; software systems for intentional communities; mindfulness and sustainability practices; permaculture;
For the following topic proposals I will request you to find an adequate company (or several) to work with.

Analysis of Intervention opportunities to increase wellbeing in software engineers' days
(An in-depth analysis with companies or one more dedicated company, by a survey followed by interviews that go deeper into the survey questions to elicit the needs and wishes and opportunities for where & how we can increase wellbeing. Then we propose options for tailored interventions that fit the needs that have come up.
Action Research on Intervention opportunities to increase wellbeing in software engineers' days
(A survey on wellbeing and resilience and then evaluate a proposed intervention developed based on the actual needs, and then evaluate the tailored solution. This requires company employees to be willing to participate.)
Biohacking amongst software engineers - how to be and perform well
(Evaluation of an intervention, to see which routines make a difference - we would need a set of a dozen people that do the intervention and a set of a dozen who don’t (but still fill out a couple of surveys for us) such that we can compare the data)
Use of recreational activities and substances in software engineering
(Survey research across several companies and a wider network of practitioners - you may need a good network for this one to have people respond.)
Mental wellbeing in software engineering companies - intention and reality
(Getting an overview of what is there, this calls for a wider survey, to check what companies are putting out there versus what the employees get.)

Examples of recently supervised theses:

Benefits of breathing practices for computer workers (2021, MSc thesis and BSc thesis)
The impact of software systems on sustainability (2021, BSc theses)
Designing a customer CRM system for the home automation industry (2016, Master's Thesis)
SE4S Project Vision Diagramming Toolkit: Build your vision (2016, Master's Thesis)
User Interface Design, Database Connectivity and Security in Fortified Cards (2016, Master's Thesis)
QR codes and communication between Android and unit testing in Fortified Cards (2016, Master's Thesis)
Protomaker - three dimensional CNC Platform (2016, Master's Thesis)
Laser Printhead Platform for Protomaker (2016, Master's Thesis)
Weighing environmental impacts in software distribution systems (2015, Master's Thesis)

Lucas Gren

lucas.gren@gu.se

Research interests: Collective Intelligence, Social Psychology, Agile Development Processes, Statistical Methods
(all in the context of Empirical Software Engineering)

Suggestions of thesis topics 2022:

Leader prototypicality in software engineering: are more typical team members more effective leaders?
Using a short team maturity measurement on a large scale at Volvo Cars.
Measurements of agile transformations in the automotive industry.
How to replicate SE experiments with tiny sample sizes: A Bayesian approach.
Sample size recommendations in SE studies.
Team maturity and planning effectiveness in automotive software development.
Interpersonal conflict in software development.
Personality or social identity: Which one is the best predictor of agile team performance?

Examples of supervised theses that resulted in a publication:

Nilson M., Antinyan V., and Gren L. "Do internal software quality tools measure validated metrics?" In Proceedings of the 3rd International Workshop on Managing Quality in Agile and Rapid Software Development Processes (QuASD 2019), Barcelona, Spain, Nov 27, 2019, DOI: 10.1007/978-3-030-35333-9_50
Mellblom E., Arason I., Gren L., and Torkar R. “The Connection Between Burnout and Personality Types in Software Developers” IEEE Software 36(5), 2019, DOI: 10.1109/MS.2019.2924769.
Bäckevik A., Tholén E., and Gren L. “Social Identity in Software Development” In Proceedings of the 12th International Workshop on Cooperative and Human Aspects of Software Engineering (CHASE), Montréal, Canada, May 27, 2019.
Al-Sabbagh K., and Gren L. “The Connections Between Group Maturity, Software Development Velocity and Planning Effectiveness,” Journal of Software: Evolution and Process, 30(1), 2018, DOI:10.1002/smr.1896.
Chronis K., and Gren L. “Agility measurements mismatch: A validation study on three agile team assessments in software engineering” In Proceedings of the 17th International Conference on Agile Software Development (XP), Edinburgh, United Kingdom, May 24-27, 2016. DOI: 10.1007/978-3-319-33515-5_2

Daniel Strüber

danstru@chalmers.se

Research interests: Model-Driven Engineering, AI Engineering, Search-Based Software Engineering, Software Product Lines

Open thesis topics for VT2022:

Smart genetic operators for search-based software engineering
A DSL for the declarative specification of model constraints (with IAV Automotive Engineering)
Debugging support for pattern matching tools in the context of model-driven engineering (with IAV Automotive Engineering)
Predicting software defects with machine learning: what makes a good project for cross-project prediction?
Taming proprietary formats of mined software models from GitHub: AI-supported parsing
A model-driven infrastructure for bridging conversational AI platforms
Feel free to contact me for additional topics!

Examples of recently supervised theses:

Generating Mutation Operators for a Search-Based Model-Driven Implementation of the Next Release Problem (thesis available)
A Maturity Assessment Framework for Conversational AI Development Platforms (paper at SAC 2021 available)
Feature-Oriented Defect Prediction (paper at SPLC 2020 available)
Clone Detection for Model Transformation Languages (paper in SoSyM, 2019 available)