Skip to content

Vis4Sense/student-projects

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Student Projects

Repository for all the student projects, including UG/PGT final projects and intern projects.

How the project works

  • While each student will have his/her own project, I encourage all students to work togeher as much as possible to share the common work such as data collection.
  • You will need to be proactive, especially if you want a good grade:
    • I will provide general directions, but not step-by-step instructions (you will need to figure these out);
    • Such as finding relevant papers, coming up with new design idea, and exploring implementation options;
    • I am always here to help if you have any question or difficulty: ask as soon as you can if something is not clear or you got stuck. Don't wait for the next meeting.
  • Have a look of the previous student projects:
    • Someone has worked on your topic already; you don't need to start from scratch.
  • We will have weekly meetings:
    • In person or online;
    • We will go through each project at the meeting (we may need to split into two groups);
    • You need to update your progress since last meeting and discuss any questions/difficulties;
    • Set the tasks before next meeting.

Project topics

  • Please see the background and possible ideas on my website.
  • We will need to further narrow down the scope for your project:
    • I will help you do this;
    • We will define the work for each student if there are more than one working on the same project.
  • You will get 80+ mark if the project results can be published in a good conference or journal;
    • This is not a requirement and it will need more work than projects not trying to publish.
    • If you want to do this, please read at least one paper from beginning to end (listed on the project idea webpage).
      • This will take a long time (at least a few days). There will be many things you never heard of and not explained in the paper, so you need a lot of extra research to understand them. The worst part is that this is often recursive.
    • Figure out what the research question is, and narrow down yours in a similar way;
    • Find a target paper (whihc is probably not be first you read)
      • This the baseline for your comparison, so you can show the improvement brought by your work;
      • Copy the process from requirements, design, implementation, to evaluation for your project. This is the way usually how these are done.

Ethics

  1. Preliminary ethics form: to decide if you need to submit the full ethics form
  2. Full eithics form (most of you will need to): for this you will need
    • recruitment materials
    • information sheet
    • consent form
    • privacy notice
  3. Data Management Plan (DMP)
  4. More details on the webpage.

Design

Paper prototype/wireframe tool:

  • Pencil and paper
  • Low fidelity: PowerPoint or Balsamiq https://balsamiq.com/
  • High fidelity: Figma (recommended) https://www.figma.com
    • Free (until Adobe starts to charge for it)
    • Web-based: easy to share with other, and allow co-editing and comments
    • Support advanced feature such as 'interaction mockup': user can click a box (that looks like a button) and a new screen is displayed (by loading another wireframe)
    • The official tutoral and there are also many on YouTube like this one.

Evaluation/Feedback

When to get feedback?

  • After the completion of the paper prototype/wireframe (less formal)
  • After the completion of the MVP (less formal)
  • After the completion of the final protoytpe (more formal)

How to get feedback and evaluate

  • It is not that different from the requirement gathering, except user would use what you produced.
  • For the wireframe, you need to help user imaging how the app would work
    • Figma 'interaction mockup' will be useful here
  • For MVP, help participants focus on the core functionalities, not the missing/less important ones (such as user account)
  • For the final prototype, it will be a proper evaluation (vs. feedback)
    • More participants: 5+ (10+ for statistical analysis)
    • Clearly defined task(s): maybe more complex so takes longer to complete
    • More detailed notes/recording: may record the session, including interviews
    • More detailed analysis afterwards: qualitative (such as thematic analysis) and/or quantitative (such as t-test or anova analysis).