Exercise II

Due on Friday, October 30, 2020 at 11:59pm

Problem Statement

Now you have finished your first exercise. Let us do a simple validation. Refer to 04-Validation.pdf for the details. For this validation process, the Domain Situation and Algorithm components are not needed. As a result, what you should do is validating Data/Task Abstraction and Visual Encoding/Interaction Idiom.

Here is a list you can do for a simple and straightforward validation.

• Revise your Exercise I and produce a set of questions, answers, and the visuals that answer the questions.
• Design a set of questions for evaluating the effectiveness, expressiveness, etc. for Data/Task Abstraction and another set of questions for Visual Encoding/Interaction Idiom. These should be questions from which you believe you should be able to understand more. For example, here is an example for evaluating an implementation of the well-known Vigenere cipher. A rather complete discussion of this cipher can be found here and the software is available here. Please review these materials as a simple guide for your question design.
• Invite ten or more MS students to participate this validation and evaluation activity. These participants should not be those students who have taken or are taking this course. First, brief them what this project is and ensure that they are willing to help you. There is no domain situation issues here because snowfall records dataset is nearly a common sense in this area. Explain the dataset, questions in your study to your participants, and presentation either via e-mail or fact to face conversation.
• Send your visuals and questions to your participants and ask them to fill out the form (of questions) and return it to you in a few days, say in 2 to 3 days. Note that it would be better to do this in an anonymous way. That is, no name or any identification information should appear on the form. Consequently, avoid the use e-mail as a mean for returning the form. Having the filled out form in your mailbox (if you have one) may be a better idea. Or, you may use a web form.
• Finally, analyze the returned data and have your findings clearly stated and reported in your report.

My Expectations

The following is a list of my expectations:

• A good set of questions that can help you understand the effectivness and expressiveness of the Data/Task Abstraction component.
• A good set of questions that can help you understand the effectiveness and epressiveness of the Visual Encoding/Interaction Idiom component.
• The process you distribute your question forms and the way you used to collect the answers.
• The way you process the collected data. This is something you should also clearly document in your report.
• After your extensive analysis, report your findings, the problems or issues in your original visuals, how will you correct these issues, etc. I hope to see an honest and thorough study, and an assessment of your work.

Submission

1. Each one of you build a web page to present your work.
2. Also prepare a set of slides in case you may be picked for presentation.
3. Send me the link to your work by e-mail before the due date and time. Make sure you will set the permission of each file used in your report properly. If I cannot access to your report, I cannot grade your work and, as a result, you get a zero.
4. The subject line of your e-mail should be as follows:

   [CS5631] Exercise II Submisson (Your Full Name)
   

5. I will e-mail you my comments and your grade.

Grading

My grading policy is very simple. A few key points are listed below:

• The quality of your questions in the evaluation form.
• Have you addressed all the issues that are important in a validation and evaluation process?
• How extensive and thorough your analysis of the results collected from the forms?
• Have you done a good presentation on overview, trend, similarity, correlation, heirarchy structures, etc.?
• Are the techniques you used in the analysis correct? Note that just computing the mean, standard deviation, median, maximum, minimum, etc. of the answer of each question is not enough for you to understand more about your work. Even though your sample may be too small, you still can use some statistical analysis to study the answers.

I will assign EXCELLENT, GOOD, FAIR and FAIL to your work. Your grade will be sent to you by e-mail with my comments.