A major component of this course will be the execution of a term project. Potential project types include the design and demonstration of a new visual analytics tool, a case study involving visual analytics of actual data for a specific purpose, or the construction of an interactive tool for a novel type of data analysis and exploration. In some cases, projects will have a client who influences project goals and requirements. In other cases, one of the course instructors will serve as the project client. Students may choose a project from a list of suggested projects (see below) or propose one of their own ideas.
Projects will generally be done by teams of 1 to 2 students. Projects will naturally be expected to require an effort proportional to the number of team members. All members of a project team will receive the same grade on the project, except in extraordinary circumstances.
Projects could also be further developed into research papers and be submitted to any of the conferences at IEEE VIS, with a submission deadline on March 31st, 2014. Students who are interested in this should contact the instructor.
Suggested Project Topics
Software and Libraries
Projects will be structured as a sequence of phases. Completion of one phase is not required for initiation of the next. In many cases, it will benefit you to be working on multiple phases at the same time. For instance, you might be working on the implementation of an initial prototype concurrently with searching and reviewing the relevant literature.
Deadlines and Grade Contribution
The below table summarizes the different phases with their deadlines and grade contributions. Each phase of the project will make an individual contribution to your grade. If a phase is missed, that portion of the grade will be a zero. You should consider this a compelling reason to start your project early and work steadily throughout the semester, rather than making a grand push at the end of the semester.
Each phase should be submitted to the instructor (Prof. Elmqvist) via email.
Describe your plans to meet these requirements in a proposal of approximately 1-3 pages. Give a five to ten minute summary of your proposal to the class. Your proposal should describe:
Here is a list of sample proposals:
Review the literature describing other approaches to problems similar to yours. Include a description of the problem that your project addresses. Describe how the approach taken in related work is similar to and different from yours. Include complete references for all papers cited. Your literature review will become part of your final paper. Your literature review should be approximately 1000 words. Correct spelling and grammar count, so check them before you hand anything in. Appropriate citation to Peer Reviewed literature is necessary.
After having defined the problem and found what other people have done in this field, now it is time for you to design the new visual analytics idea you are going to use to solve your problem. The design document is at least one page (submitted as a PDF) of your thoughts on the visual design, analytics methods, your design decisions, your intended audience, other work you are building on, and so on. Examples of information to include in your design document are
Please feel free to include any sketches you made as part of the design process. The design document does not have to be a perfect and polished piece of work: it should be a snapshot of your creative processes at this point in the project.
Your initial implementation goal is to produce any visualization of your data. Your representation and interaction techniques will not be nearly as refined as you plan to make them, but you should be able to read in a data set and produce an image showing some representation of the quantities of interest and some analysis. Submit an image electronically (URL is fine) and short description (about a page) of what the image shows and how it was produced. Also submit a description of the capabilities at alpha release.
By beta release, your project should be a complete prototype with all of the functionality that you have proposed. Your beta release should be accompanied by a short description of bugs you plan to fix and enhancements you plan to make. The third phase of the development effort will center on refinements of your prototype. Schedule a demo and meeting with your client and a course instructor in order to present your beta release and discuss a revision plan. Write a technical description of your implementation (1-2 pages). Include any images you want, but there is no need to include any source code.
A complete draft of your paper (see below), including figures and images, must be submitted in advance of the final paper deadline. This draft will not be graded. The course instructors will serve as reviewers for these papers and make suggestions as to how they might be improved. The more you include in your draft, the more the instructors will be able to help you. You may submit earlier, not necessarily complete, drafts of your paper if you would like feedback earlier in the writing process. Submit the paper as an e-mail to the instructor!
Paper formats: http://www.cs.sfu.ca/~vis/Track/tvcg.html
Write a 5000 word technical paper describing your project in the style of an IEEE VAST, Visualization, or Information Visualization Technical Papers or Case Studies (other formats may be acceptable with pre-approval). It should be a complete paper that is as strong and polished as you can make it. Aim for something that you believe is ready for submission to a conference or journal. Sections you should plan to include are (note that these are not absolute, feel free to use your own disposition):
Your paper should include figures and images as appropriate. Correct spelling and grammar count in all submitted work, so check them before you hand anything in.
Paper formats: http://www.cs.sfu.ca/~vis/Track/tvcg.html
Refine and enhance your prototype into a final release. You should attempt to respond to all requests of your client and the course instructors. Your final release should include:
We will simulate a conference reviewing process in this course. Upon all students submitting their final papers, the instructor will serve as the PC chair and assign each student up to three (3) papers to review. You will be given a week to read the paper of your fellow student and write a fair and objective review (using a review form provided to you). The reviewing process will be single-blind, i.e. your name will not be disclosed to the authors of the paper. When you are done, the instructor will collect all reviews for each paper, and will write a summary review. You will be given all three reviews for your paper, as well as this summary.
Like for any peer-review process, confidentiality is very important. Students are not allowed to discuss the contents of different papers between each other! Be objective and try to do as good a review as possible---remember that the quality of your reviews will also be part of your grade!
The reviews will serve as supporting material in giving you a final grade for your paper, but the instructor will naturally read and grade the papers independently of this.
See the guidelines on HowToReviewPapers for more detail.
Prepare and present a 15 minute presentation of your project that will be held in class in the last week of the course (before finals week). Your presentation should be professional enough to give at a technical conference (e.g. organized approach, prepared slides, a short demo or video if appropriate). Invite your client to your presentation and/or schedule a private demo.