Syllabus 

 Course Material 

 Assignments 

 Class Project

 Online Resources 

 Home

The following is a final checklist of what you need to turn in for your project. Please note that these are general requirements and in specific cases, not all of these requirements may apply. If you are not sure about what to include, please consult with me before your submission. Also, if you think that installation of your system may require non-trivial or unusual steps, you can arrange for a demonstration of your system.

Research Papers

Research Papers will be evaluated based on thoroughness, soundness, clarity and organization. The overall structure of the paper is up to you, but you must have the following sections in addition to the main body of the paper:

1. Abstract: This is a short synopsis of the main points of the paper. This should be 200-300 words, and should appear along with the title and your name, ID number, and email, on the first page. The rest of the paper should start on page 2.

2. Conclusion: Summarize your conclusions and findings. You conclusion section must include your own original thoughts and discussion of your findings.

3. References: This is a list of references that you have used in doing your research and throughout your paper. The references should be numbered and the number for the reference should appear in the appropriate places in the text of the paper where the reference was used (it is not enough to list a bibliography at the end of the paper without actually using any references within the body of the paper). You can look at any of the papers assigned for reading in the class for acceptable uses of references. URL references should only be used for referring to specific system Web sites and not as a way to reference published work.

It is highly recommended that you create appropriate sections and subsections to create an easily understandable organization corresponding to the topics covered. Your final paper should be submitted electronically in either MS Word, PDF, or Postscript formats. Submissions in other formats will not be accepted.

Implementation Projects

The Implementation projects will be evaluated based on significance, design, correctness, documentation, and appropriate evaluation/testing. You will need to electronically submit a compressed file containing your project distribution files and documentation. Your project documentation should contain the following components:

1. A detailed description of your system (including specific techniques and algorithms you used), and the interaction between the components (make references to code segments, modules, methods, functions, etc. as necessary). Your write up should also include an evaluation of your system demonstrating its correctness and functionality. If you used any outside sources in your implementation, please clearly indicate which sources, and how and where they were used.

2. Complete (actual) sample runs of your program with description, illustrating how your system works, along with any intermediate input or output used for the sample run.

Your project distribution files should contain the following:

1. Complete source code (be sure that your source code is fully documented and easy to read).

2. Binary files (e.g., executables, DLLs, Class files) or other components necessary to run your program.

3. Readme file containing instructions on how to compile, install, and/or run your program.

4. Project report/documentation, as described above.

5. Any test data used for evaluation of your system.

6. If your application is CGI-based or otherwise has a server component, please provide a URL for a demo version of your system.

Data Analysis Projects

Data analysis projects will be evaluated based on significance, design (specification of the data mining problems to be used), thoroughness (completeness of the analyses and DM tasks performed on the data, as well as well as thorough discussion of the findings and conclusions),  and documentation. You will need to electronically submit a compressed file containing your project distribution files and your project report. Your project report should contain the following components:

1. An executive summary summarizing the project goals, methods used, and conclusions.

2. The main body of the report providing details on the data set used, design decisions, description of how the KDD cycle was applied, tools used to perform specific DM tasks, a detailed description of the results of data mining, and conclusions.

Your project distribution files should contain the project report (as described above), any code written to perform specific tasks or to prepare the data, the full data set (or a sufficiently descriptive sample of the data) used for the project.