1) - UH
Group1 ML Project Topics 2009
Your Topic is approved (but submit topic update by March 6, 2008)
1. Sameer Gottipati: How do you plan to evaluate that the proposed Bayesian approach really works well/badly??
2. Cyril Harris: Need a paragraph how the very specific problem you solve fits into the “general picture”!
3. Vitek Karihaloo
4. Chaitali Kulkarni: Q-Learning topic is okay (the other project isn’t); try to come up with a good case study in the next 10 days how the proposed system will be tested!
5. Sasi Pitchaimalai
6. Khai Tran: Define more clearly the image processing problem solved and what a background subtraction model is and what it is revised from!
7. Anshul Verma: “Image Boundary” theme is fine; continue with this theme!
8. Tarun Wadhawan
Your project theme is fine but there are some issues
1. Gaurav Chandra: What datasets will be used; how will the quality of the proposed approach assessed?
2. Duc Duang: Your proposal is very general. Can you a more specific proposal? How will the explored techniques be evaluated?
3. Emil Ismailov: Need project title! What is the exact problem the proposed approach will solve?
4. Geory Golovko: Explain in more detail how machine learning will be used in the proposed project!
5. Nupur Kulkarni: Either the Handwritten Digit Recognition or the Aircraft Operations theme are fine. Pick one as soon as possible!! Address how the proposed techniques will be evaluated and what datasets will be used!
6. Javed Ahmed Mohammed: Forward Backpropagation scheme is fine but submit a more detailed description of what will be exactly done and how it will be tested! What is done by the tool you use and what is done by you?
7. Nivedita Rai: What datasets will be used; how will the quality of the proposed approach assessed?
8. Philip Trevino: The overall theme is good but your project description is quite general; I recommend that you propose a more specific subtopic of the more general topic you propose in about a week (e.g. you only look at numbers/numbers with noise in specific set of images, and generalize the approach if there is any time left.)
9. Song Wei: What are operators/states in the proposed RL approach? How do you cope with congestion, if everybody uses the same strategy? Try to design a good case study to test the proposed approach soon!
10. Karthik Vermuri (change of topic!: Questions: What texts will be used to evaluate you system; what benchmark will be used; how do you evaluate that the Bayesian classifier does a good job; what information (e.g. word counts,…,?) will be used to classify text. Bring a hard-copy of a more detailed project specification to the lecture on March 9, 2009 and also mail the softcopy to Dr. Eick!
Remark: all students listed in this file form Group1, the students not listed form Group 2.
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.