1. What dataset will you use? Where is it from? Provide a URL for the dataset if possible. If not, get me the dataset another way before the due date.
2. What question(s) do you hope to answer with the data?
3. What data mining approach(es) will you use? Is the problem supervised or unsupervised? Classification or regression? Will you compare different classification approaches or will you attempt to answer a very specific question from the data?
4. How will you determine success or failure? How will you evaluate the quality of the results you will get from your models?
5. What other data scientists have mined this data? Hand in at least one journal article or whitepaper describing an approach used to mine the data. If you've found something and you're not sure it qualifies as a journal article or whitepaper, ask me before handing in.