research proposal on data science

Select language

Data Science

Part 1: project proposal.

The process described on this page follows after the process in the step Getting started . The duration of the Research Project starts now. This part is called 'Data Science Project Proposal' in Osiris (INFOMDASC1).

Follow the instructions on this page to complete Part 1: Project proposal.

Part 1 comprises 15 EC and focuses on the project proposal. Although the name proposal may be a little misleading, this part is more than simply a proposal. It is intended to include all this work that is necessary before devising the actual solution to a research problem, for example the literature study. Its minimum duration is 10 weeks of full-time work. After you done this, Part 1 will give a Go/No-go decision towards Part 2.

A project proposal consists of:

An introduction to the problem the Research Project deals with . In particular, it is explained here why the topic is interesting, why it is important, who would be interested in it, why it is technically challenging to be a Research Project, and why it is now the time for such research work. The produced text will become the introduction of the thesis document.
A literature study so that one obtains a complete overview of the existing works, before starting to develop the solution to the problem at hand . This should not be simply references to published work but should explain the approaches in a complete and comprehensive way, such that someone who reads this part of the document will have a good understanding of what related studies and solutions have been done and how they work. Note that this part will become the 'Related Work section' in the final thesis.
A formulation of the research questions and a formal specification of the problem statement . This will also help in the subsequent steps to stay focused on the actual challenge at hand. This part can be later become a problem statement section in the final thesis document.
A plan of how the research will be conducted I.e., how the solution will be developed, what datasets will be used, and how the outcomes will be experimentally evaluated.
A completed Ethics and Privacy Scan (see below for details). Failing to complete this scan may result in an automatic falling (no-go) decision of Part 1 of the Research Project.

Instructions Ethics and Privacy Quick Scan (mandatory)

Ethics and privacy quick scan (mandatory).

By the end of the Part 1 phase of your project, you must submit the Ethics and Privacy Quick Scan. This document helps you identify any ethics or privacy issues with your project and to address them on time, before starting your main Research Project work in Part 2.

Submitting the Ethics and Privacy Quick Scan before the start of Part 2 of your project is mandatory.

When you submit the Qualtrics version of the scan, you and your project supervisor/first examiner will receive the ethics and privacy scan outcome report by email and as a PDF. In case no issues have been found, or if your project supervisor/first examiner already has ethical approval for their line of research in which the master thesis will take place, then the Part 2 project phase can start from an ethics and privacy perspective.

Keep the Quick Scan report

If the Quick Scan does not provide any warnings regarding your project you can add “ The Ethics and Privacy Quick Scan of the Utrecht University Research Institute of Information and Computing Sciences classified this research as low-risk with no fuller ethics review or privacy assessment required ” to your Part 1 document, as well as your thesis manuscript and eventual publications.

Please also attach the anonymized Quick Scan report as an Annex to Part 1 of your thesis. For more details, please see 'Guidance'-section of the Ethics and Privacy website .

Tips to complete the Ethics and Privacy Quick Scan:

Read the Word version of the Quick Scan. Do this early, as some participants/data/topics raise more ethical and privacy concerns than others. Consider beforehand the possible issues and adapt your research questions and research designs if needed, in agreement with your supervisor.
Once you and your project supervisor/first examiner are satisfied with the outcome, submit the Quick Scan using the Qualtrics version, also found on the aforementioned Ethics and Privacy website.

All the above you have to describe in a document that you deliver to the project supervisor/first examiner. They will provide feedback, evaluate it and based on the evaluation provide a Go/No Go decision. This decision will be entered in Osiris. The assessment is based on the scientific quality, the writing skills, and the planning.

Experiment and evaluation type project

The above description assumes a type of a project that is developing a solution to a research problem. In some cases, a thesis may be of a different type. It may be an “experiment and evaluation” type, or a study. In these cases, a number of existing works from the literature are studied, evaluated and compared to reach to some research conclusions. In these cases, the exact format of the sub-part 1 above and of the expected Research Project in general will be determined by the project supervisor/first examiner.

An  example assessment form of Part 1 with more detailed criteria is available. Please use this form only as a discussion piece and do not send in paper or scanned forms.

Go or No Go decision

There are three options for the Go/No Go decision:

Sufficient: means that you automatically continue with part 2.
Insufficient with re-take exam : means that your project supervisor/first examiner explains which parts need improvement. A date to re-evaluate part 1 will be set. After re-evaluation, your (first) supervisor/first examiner will send the updated assessment to the Student Desk.
Insufficient without re-take exam (fail) : means that you have failed on the above-mentioned learning goals and have to abort the research project. Your project supervisor/first examiner will report his/her findings to the programme director who will invite you and the Study Advisor. The programme board, after consulting you and possibly other people, will decide whether starting up a new project will make sense.

What do I need to do if my project gets delayed?

Students with a project from 1 september 2024:.

This is what you need to do if you foresee a delay of your Research Project or extension/addition to the project is necessary.

The protocol*

The student and examiners need to finish the Research Project before the in Osiris Zaak specified end date. The end date is the last date by which the final grade is determined. The end date is based on full-time study.
If the end date cannot be met, the student, first and second examiner agree on a new end date. This new end date will be passed on to the Board of Examiners by the student via Osiris Student > ‘My Cases’> ‘Start Case’ > ‘Request to the Board of Examiners’ > ‘request type ‘New end date thesis project’ . This needs to happen before the initial end date is reached. Valid reasons for agreeing on a new end date can be both personal circumstances and research-related circumstances.
The student and examiners can impose an examination on the agreed end date. In the case the other party does not agree with this, they can turn to the programme leader. A student who due to circumstances beyond their control cannot be present during examination can request the Board of Examiners for a special testing provision.

The student and/or examiners can turn to the Board of Examiners in cases of disagreement on the implementation of this protocol or other conflicts not covered by this protocol. In these cases, the Board of Examiners decides in line with the spirit of this protocol.

*This protocol is translated from the Dutch version in the EER/OER and no rights can be derived from any errors in translation.

Students with a Research Project from before 1 September 2024:

This is what you need to do if you, due to circumstances beyond your control, foresee a delay of your Research Project in Part 1 or Part 2.

The procedure:

Discuss this first with your supervisor(s). If all agree a new realistic end date will be set for the Research Project.
After that, contact the Study Advisor and the programme coordinator and ask for consent to determine a new end date for your thesis.
Apply for an extension of the research project deadline for Part 1 or Part 2 to the Board of Examiners via Osiris Student > ‘My Cases’> ‘Start Case’ > ‘Request to the Board of Examiners’ > ‘request type 7 'Delay of research or thesis project'.

What information is needed for the application form:

A statement from the Study Advisor
A copy of an email in which the supervisors support the request for a deadline extension
A proposed new deadline
Short statement to support your request

Colloquium (INFOMCDASC)

Once Part 1 of the Research Project has been completed, a colloquium needs to be held .

Proceed to 'Colloquium' ⋙

Proceed to 'part 2: completing your thesis' ⋙, follow utrecht university.

Utrecht University Heidelberglaan 8 3584 CS Utrecht The Netherlands Tel. +31 (0)30 253 35 50

Data science project proposals

Are you wondering when you should write a data science project proposal document? Or maybe you are wondering what content you should include in a data science project proposal? Well either way, you are in the right place! In this article we tell you everything you need to know about writing data science project proposals.

First, we talk about what a project proposal is and what the purpose of writing a project proposal document is. After that, we talk about when you should write a project proposal document. We follow this up with a description of what content should and should not be included in a data science project proposal document. Finally, we provide tips for writing a strong data science project proposal document.

What is a data science project proposal?

What is a data science project proposal? A data science project proposal is a document that is written when you want to propose a new data science project idea. This document should include information about the problem that you want to solve and why it is important to solve that problem, among other details.

Why should you write a data science project proposal?

Why should you write a data science project proposal? The main reason to write a data science project proposal is to ensure that you are aligned with your stakeholders on what your team should be working on. Specifically, a project proposal document should be used to align on what problem will be solved and what constraints the solution needs to adhere to. It should also be used to align on timing and whether now is the right time to work on a given project.

It is important to put your vision for the project down on paper so that it can be reviewed by your stakeholders. Putting your thoughts in writing makes it easier for stakeholders to give asynchronous feedback. It is also less common for miscommunications and misunderstandings to happen when there is a written document to reference. As an added benefit, a project proposal document will serve as a key piece of documentation that future team members can reference to understand why a project was prioritized.

When should you write a data science project proposal?

When should you write a data science project proposal? You should create a project proposal document early in the project lifecycle. Creating a proposal document should be one of the very first steps you take when starting a new project. The only activity you might perform before creating a proposal is running an impact sizing exercise to estimate the scale of potential impact.

You should make sure that you do not invest a lot of effort into a project before aligning on the project proposal. That project proposals can be rejected and or put on the back burner for later. You do not want to invest a lot of effort into a project just to learn that your stakeholders are not aligned on the need to work on the project.

What should be in a data science project proposal?

What should be in a data science project proposal? Here are some of the main topics that should be covered in a data science project proposal.

Problem . The first thing you should include in your project proposal document is a description of the problem that you intend to solve by working on this project. It is important to put this in writing to ensure that everyone is aligned on exactly what problem will be solved.
Reason for solving this problem now . The next thing you should include in a project proposal document is the reason that you should solve the problem now. There are likely to be many different business problems that your team could be working on at any given time, so it is important to have a justification for why now is the right time to solve a given problem.
Constraints . The next section you should include in your data science project proposal is a list of constraints that your solution needs to adhere to. This should include both business constraints and technical constraints if possible. If you do not have enough familiarity with the technologies you will be using to understand the technical constraints they will impose without doing some exploration, you can stick with just the business constraints your solution needs to meet.
Goals . The next thing you should include in the project proposal is a list of goals for the project. These goals should describe the criteria that needs to be met in order for the project to be considered a success.
Non-goals . It is just as important to include information on what will not be included in the scope of a project as it is to include information on what will be included in the scope of a project. Make sure to include details on edge cases that you will not be tackling a part of the project.
Business metrics and impact sizing (if applicable) . Finally, you should include information about the business metrics that you intend to move with this project. If there is not a clear business metric that the project will move, you should reconsider whether you should be working on that project. If you are able to give an estimate of the size of the impact you might expect a project to have on a given metric, you should also include that information in the proposal.

What should not be in a data science project proposal?

What should not be included in a data science project proposal? Here are some examples of information that should not be included in a data science project proposal .

Implementation details . Project proposals should focus on the nature of the problem that needs to be solved and the constraints that the solution needs to adhere to. They should not go into detail on what solution will be built or how the solution will be implemented. Details about how the solution will be implemented should be saved for a technical design document that should be created later in the project lifecycle. You should not have all of the details necessary to describe the solution that you will build this early in the project lifecycle.

Tips for creating a strong project proposal?

How do you write a strong project proposal that will drive alignment with your stakeholders? Here are some traits that characterize a strong project proposal document.

Succinct . You should aim to make your project proposal document succinct and only include details that are necessary. There are multiple reasons for this. For one, it will make it easier for stakeholders to skim your document. This will increase the likelihood that any given person will actually read through your document.
Opinionated . A project proposal document should be opinionated. It should state clearly what should and what should not be done as part of the project. The more opinionated the document is, the more likely that the alignment that is achieved in the proposal stage will carry throughout the rest of the project. Stakeholders will understand exactly what the team is and is not committed to up front.
Accessible . A project proposal document should be written for a broader audience that contains both technical and non-technical stakeholders. It should not contain specific technical terminology or references that would make it inaccessible for a less technical audience. You should always aim to get feedback on your document from a wide variety of stakeholders with different backgrounds and experiences, so you need to avoid writing the document in a way that is only accessible to a small subset of people.