Choosing a topic (due November 3):

Please choose a topic that focuses on a research problem that uses probability models. The probability models should either be those covered in this course (e.g. Markov chains, Poisson processes, queues), or extensions of those coverered in this course (e.g. renewal theory, more advanced queues).

You can not cover material covered in lecture.

Your project will consist of learning about the topic and presenting what you have learned to the class. You are not expected to build or design anything.

You should work in a group of two students. If you are looking for another student to work with, you might want to post to the CS 278 MessageBoard.

By November 3, your group of two should post information about your proposed project on the course wiki. Please include the names of the students in your group, a project title, and a short (one paragraph) description of your topic. The description need not describe everything you will investigate, but it should be enough to give me an idea of what will be included and what is outside the bounds of your interest.

Learning about your topic:

You should find and read information related to your topic. This information should be composed of a few independent sources. You can use any source (paper or electronic) that you deem respectable. Here are a few research journals that I believe are generally reliable, but there are many other journals in other areas of computer science, mathematics, statistics, engineering, economics, and business that include papers that use probability models.

• IEEE Transactions on Automatic Control
• IEEE Transactions on Communications
• IEEE/ACM Transactions on Computational Biology and Bioinformatics
• IEEE Transactions on Computers
• IEEE Transactions on Image Processing
• IEEE Transactions on Information Theory
• IEEE Transactions on Mobile Computing
• IEEE Transactions on Multimedia
• IEEE/ACM Transactions on Networking
• IEEE Transactions on Network and Service Management
• IEEE Transactions on Pattern Analysis and Machine Intelligence
• IEEE Transactions on Signal Processing
• IEEE Transactions on Vehicular Technology
• IEEE Transactions on Wireless Communications
• Journal of the ACM
• ACM Transactions on Algorithms
• ACM Transactions on Computer Systems
• ACM Transactions on Database Systems
• ACM Transactions on Embedded Computing Systems
• ACM Transactions on Internet Technology
• ACM Transactions on Modeling and Computer Simulation
• ACM Transactions on Multimedia Computing, Communications, and Applications
• ACM Transactions on the Web
  

A reasonable amount of information would probably be 3 articles from these journals, or the equivalent amount of information from other sources. The papers should pertain to very similar problems, but use different probability models (or use similar models in different ways). Your goal is to find several papers that address similar problems and to understand how and why these use different models. It is ok if some of the papers use closed-form analysis and some use simulation.

You will have two deliverables: a presentation to the class and a written report.

Presentation:

Your presentation will take place on one of the following days:

• Tuesday, December 1 (2:00-3:20)
• Thursday, December 3 (2:00-3:20)
• Thursday, December 10 (1:30-3:30)

Your group should sign up for a time slot on the course wiki. Each time slot will be 15 minutes. (Groups of 1 will have 7.5 minutes and groups of 3 will have 22.5 minutes.) You are expected to attend all days regardless of which day you are presenting.

Your talk should use a set of computer projected slides. You may use either ppt or pdf formats. You may use your own laptop or you may use the computer in the classroom.

You should practice your talk to learn the timing, as I will cut you off when you have used 15 minutes. Although 15 minutes might seem like a long time, you will not have time to present everything you have learned from the papers you have read. You should focus on the following questions:

1. What is the general problem that you are addressing, e.g. allocation of bandwidth in networks? (perhaps 2 minutes)
2. What is the specific goal of the model, e.g. determine the mean revenue earned by a specific bandwidth allocation algorithm? (perhaps 3 minutes)
3. What are the models used in each of the papers you read, and why did they choose different models? (perhaps 8 minutes)
4. What are the limitations of these models? (perhaps 2 minutes)

You should decide what you think is the most important information for your classmates.

YOUR LAST SLIDE MUST BE A BIBLIOGRAPHY OF YOUR INFORMATION SOURCES. Please include the following information for each:

1. Author of the document.
2. Title of the document (e.g. article title or web page title).
3. Venue where the document was published (e.g. journal, conference, magazine, publisher, name of website).
4. Where to find the document within the venue (e.g. volume number and page number, or path within webpage sitemap), if available.
5. Date written; for webpages without a date written include the date accessed.
6. URL, if available. This should be the direct URL provided by the venue, not the URL used through a search engine or abstracting service.

You should post your slides (in either ppt or pdf format) to the CS 278 Dropbox by the day you present.

Grading:

You will be graded primarily on the effectiveness of your transmission of information to your classmates. Specifically, I will assign scores for preparation (how well did you demonstrate that you know the material), content (which material did you include in your presentation), presentation (how well did you communicate this information to your classmates), and length (how closely did you match the 15 minute target). In addition, you will receive comments from me and from your classmates.