A student can find a thesis project internally at the IT Department or externally at a company or organisation. 

The Department of Information Technology conducts research in the following research areas: image analysis, computer systems, optimisation, computing education research, parallel and distributed systems, control and dynamical systems, programming languages and systems, cybersecurity, semantics and verifications, data science, software engineering, embedded and real-time systems, artificial intelligence, human-machine interaction, and computational science. You can read more about it here.

If you are interested in an advertised thesis project at the IT department at the university, please get in touch with the contact person by email, phone, or in person.

If you are interested in a thesis project in another department, another university, or a company, make sure you have a CV and cover letter ready for submitting an application.

Remember that being proactive and prepared are key to being selected. Proactivity means taking initiative and anticipating what needs to be done before it's even asked of you. Don't wait for instructions; instead, actively seek ways to contribute and make a difference. Be ready to dive in with enthusiasm, knowledge and skills relevant to the project's objectives. Furthermore, think deeply about how you are a good fit for the project. Consider your unique strengths, experiences, and passions that align with the project's goals. Demonstrating your genuine interest and how you can add value will increase your chances.

Uppsala University Career Support offers a number of services, including feedback on employment letters, help with writing a CV and cover letter and career counselling. You can book an appointment with them here.

It is the responsibility of the student to find a suitable thesis project. Please start applying for projects early as responses from a Supervisor / Handledare at a company or university may take some time.

Please note that not all external projects are suitable for Masters thesis projects. First and foremost, projects should be relevant to the degree course the student is a part of. Projects primarily involving implementation (program design/coding/testing/debugging) work are not generally acceptable (although sometimes an implementation project can be made acceptable by including suitable design issues).

As a rule of thumb, not more than a third of the project should be implementation work. If the implementation presents particular challenges which are connected to the subject of the thesis project, then up to half the effort may be implementation. Also, projects where the student is continually working from instructions are also generally not acceptable as the student should carry out independent work.  To prevent this, it is advisable to find a thesis project at the university or at a company that has experience providing thesis projects for students. 

Master's Programme in Image Analysis and Machine Learning, 120 credits (TBA2M)
The project must be clearly connected to the area of image analysis, which nowadays typically brings in machine learning approaches. An obvious way to meet this requirement is to work with image data. If this obvious criterion is not met, the project description and, ultimately, the report itself, must convincingly explain the connection and relevance to the field of image analysis and machine learning. In general, it is not sufficient to work on a project which involves machine learning on data other than images. Exceptionally, and based on sufficiently strong arguments, such a project might be approved (however, this requires discussion with the course examiner prior to approval of the specification).

Furthermore, a successful project meets the requirements defined by the stated learning outcomes, available here. Most important of these requirements are:

• There are clearly defined and delimited goals of the project, outlining relevant (and clearly stated) scientific and/or technical image analysis related problems to be addressed.
• The chosen method(s) are connected to the scientific literature. It is expected that the choice of the method(s) is justified based on the image analysis problem at hand and contextualized with respect to possible alternative choices.
• The presented approach is evaluated by following a proper scientific approach. It is assessed (quantified) to which extent the presented approach meets some clearly defined evaluation criteria. This may be in the form of empirical evaluation and/or theoretical validation (e.g., as mathematical proof). The evaluation method(s) should be suitable for the chosen criteria, and be motivated as such.
• The key findings of the project are analyzed and discussed in depth, both positive and negative when applicable, in a manner that demonstrates understanding and knowledge of the subject matter.
• Whenever applicable, societal and ethical aspects of the problem, as well as of the solution, are discussed. Some remaining open problems are identified.
• The project is a result of an independent (supervised) work performed within the defined time-frame. It is summarized in a written report of a high linguistic quality, which follows a predefined format.

In particular, note that pure implementation work is not considered sufficient for a degree project. A comparative analysis of several existing methods, which are selected based on a literature study, implemented, and applied to solve a certain image analysis task, and then evaluated based on a set of suitably selected criteria, is considered to be a suitable master project.

Master's Programme in Computer Science, 120 credits (TDV2M)
The project must be clearly connected to the area of Computer Science. Computer Science is a broad area and there is a substantial overlap with other areas. In cases where the project is not within the core areas of Computer Science, the student should check with the examiner well in advance that the project is suitable and the project description must clearly explain the relevance for Computer Science.
Many project proposals concerns applications of Computer Science. Such projects often have enough Computer Science content to make them suitable as degree projects, but it is important that the Computer Science content is highlighted and that issues related to the application area are not allowed to dominate. The scientific problem must be in the area of Computer Science and not in the application area.

It must be understood that software development is not a scientific problem, which is required by the syllabus. Thus software development projects are not suitable as degree projects. Software development is allowed insofar it supports the scientific study. E.g. a project about developing new algorithms may need software development to implement the algorithms for making validation and evaluation possible. That is acceptable, but as a rule of thumb such software development may not take up more than a third of the project time. If the software development itself is the subject of the study, e.g. if the project studies code quality obtained from different software development methods, then up to half of the time may be spent on software development.

It is also required that:

• There are clearly defined and delimited goals of the project, outlining relevant scientific problems to be addressed.
• The chosen method(s) are connected to the literature. It is expected that the choice of the method(s) is justified based on the problem at hand and contextualised with respect to possible alternative choices.
• To perform a scientific validation of the presented approach to evaluate the extent to which it meets some clearly defined evaluation criteria. This may be in the form of empirical evaluation and/or theoretical validation (e.g., as mathematical proof). The evaluation method(s) should be suitable for the chosen criteria, and be motivated as such.
• To analyse and discuss in depth the key findings of the project, both positive and negative when applicable, in a manner that demonstrates understanding and knowledge of the subject matter.

Master's Programme in Embedded Systems, 120 credits (TIS2M)
The project must be clearly connected to the area of embedded systems. In cases where this is not obviously so, the report and the project description must convincingly explain the connection and relevance to embedded systems.

It is also required that:

• There are clearly defined and delimited goals of the project, outlining relevant scientific and/or technical problems to be addressed.
• The project is put into a broader context based on the existing literature of related work, which is properly cited, with key differences and similarities of the related work to the current project pointed out.
• The chosen method(s) are connected to the literature. It is expected that the choice of the method(s) is justified based on the problem at hand and contextualized with respect to possible alternative choices.
• The presented approach is evaluated by following a proper scientific approach. It is assessed to which extent the presented approach meets some clearly defined evaluation criteria. This may be in the form of empirical evaluation and/or theoretical validation (e.g., as mathematical proof). The evaluation method(s) should be suitable for the chosen criteria, and be motivated as such.
• The key findings of the project are analyzed and discussed in depth, both positive and negative findings when applicable, in a manner that demonstrates understanding and knowledge of the subject matter.
• Relevant future work and open problems remaining after the project are identified.

In particular, note that pure implementation work is not considered sufficient for a degree project. Projects that include a significant implementation part must still meet the requirements listed above.

The report must follow the univresity guidelines. Please notice that a report not following this style will be returned to the student without additional feedback, which may significantly delay the thesis revision and examination process.

Master's Programme in Data Science , 120 credits (TDA2M)
General requirements
The project must include:

• the development of new data science concepts and/or methods,
• and/or the application of existing data science methods to a problem.

It is also required that:

• The chosen method(s) are connected to the literature. For example, in an applied data science project it is expected that the choice of the method(s) is justified based on the problem at hand and contextualised with respect to possible alternative choices.
• To perform a scientific validation of the presented approach, e.g. regarding efficiency, performance, the extent to which it meets some clearly defined evaluation criteria.
• To perform a societal/ethical validation. In particular, it is mandatory to include a well-thought section of the report on ethical reflections and societal impact. If no ethical considerations/implications are relevant for the given project, this has to be explicitly mentioned.

The report must follow the univresity guidelines. Please notice that a report not following this style will be returned to the student without additional feedback, which may significantly delay the thesis revision and examination process.

Questions & Answers
Q1: what counts as a data science problem/application area?
In case of an applied project, any application area is ok as long as it requires advanced data science methods.

Q2: what counts as an advanced data science method?
Any method that would fit one or more courses in the programme outline of any of the two specialisations, with the exception of the Bachelor-level bridging courses Database Design I and Linear Algebra for Data Science, shall be considered a good fit for the thesis project.

Q3: Must the topic of the thesis be clearly connected to a course in the programme?
No, it does not have to, as long as it concerns:
the use of computational methods to extract knowledge from large and/or complex data,
and/or the theory of computational methods to extract knowledge from large and/or complex data.

Q4: What does “large and/or complex data” mean, by the way?
Any data that requires the application of advanced data science methods to solve the problem at hand. A database with 1000000 employee records is neither large nor complex if we want to list the names of the employees in alphabetical order, because we can solve this task with a simple SQL query. Learning a model to determine which employees should be fired based on the expected impact on the company’s revenues would make this data complex, and maybe even large (depending on the used methods), although this might not pass the societal/ethical validation.

Q5: If I study one specialisation, should I focus on projects that are related to the courses in my specialisation?
The degree is in data science, independently of your specialisation (which is “just” a way to help students choose a homogeneous set of courses leading to advanced topics in the specialisation area). So a student registered to one specialisation can do a thesis that is related to courses listed in the outline of the other specialisation, as long as the student has (or can acquire as part of the thesis) the needed competencies and skills.

Once you find a thesis project internally or externally, you need to submit an application approved by your Supervisor to formally start your thesis project. The application contains the following documents:

1. Application form:

• Master/Bachelor student form for applying for a thesis project here.
• Engineering students (civilingenjörer) here.

The form must be filled out electronically or typed; handwritten entries will not be accepted. Signatures can be typed, drawn, or written. The form must be in PDF format.

It is the responsibility of the student to use the right form and provide the correct information in the form, if you are not certain, please contact the Thesis Coordinator.

Please leave the section of the Subject Reviewer blank.

2. Thesis specification:
Approved by your Supervisor. Must contain the following sections. A template in Word and Overleaf is provided. Please leave the placeholder of the Subject Reviewer in order for the Subject Reviewer to fill it in.

Title
The title is preliminary, ie the report can have another title. The title should not be longer than about 12 words.

Abstract
A brief summary or overview of your proposal. It provides the reader with a concise description of the project's main purpose, research questions, methods, and conclusions. Abstracts help readers quickly determine whether a particular project or research is relevant to their interests or research needs, without having to read the full proposal. A well-written abstract should be clear, concise, and informative, and it should accurately reflect the content of the full proposal.

Background
Here you describe in what context your thesis is to be done. What prerequisites are valid, what is the goal of the project from the supervisors point of view, what is available and has been done before, under what circumstances should the work be done.

Description of the task
Here you describe in more detail the contents of the project: what should be done and what moments are included. Specifically it should be described the interesting part of the problem, how this should be analyzed and solved. Here there should be clarified that the project fulfills the requirements that the university has posed on a thesis project on this level.

Methods
What systems, tools and methods should be used. Relevant literature (it is often a part of the work to find additional literature). How results should be evaluated and documented.

Relevant courses
A list of courses that are of particularly interest for the project.

Delimitations
It is also important to specify what is not part of the project. This prevents that the thesis grows uncontrolled. You can add stuff that you can do if there are time left, also write down things that can be skipped if there are not time for them.

Time plan
Here you find the time plan for the project. When to start and how much time has been allocated for each moment. (the fine grading can vary, but no blocks bigger that about 4 weeks should be listed). The time plan should take into consideration factors such as part time work, vacations, other obligations etc. Some moments, like writing, is usually done in parallel with other tasks. A graphical plan is encouraged. Regular meetings with your reviewer should be booked.

References
One or more relevant references.

Applications are submitted to HotCRP system which can be found here. The deadline for submitting an application is 15 Dec 2023. No applications will be accepted after the deadline.

If accepted, you will be registered to the course in Ladok on the first week of the start of the period. You will also be assigned a Subject Reviewer.

Hints:

• Remember that your Subject Reviewer is going to read your proposal, write accordingly!
• Examples of previous specifications can be found here, here, and here. Don't forget to use the new templates provided for the specification.
• Examples of previous theses at the IT Department can be found here. You can also contact the Program Coordinator for exemplary thesis reports.
• Write down what is considered to be the core of the thesis, and what has potential to be the base in the analysis part of the report.
• Find out what research subject (or what course) that are closest to the project. Write the specification with that in mind.
• Do not copy/paste the project advertisement, lift parts that makes the specification interesting to the researchers in the area. This makes the specification more clear and concrete.
• We strongly recommend you avoid confidentiality agreements with companies and confidential reports. If necessary, it is recommended to use the form found here.
• If you have any questions about research ethics please contact Iordanis Kavathatzopoulos.