Tuesday 5 October
- quiz 3: DM ch4 (4.6 to end of chapter) & pg2DM (open note)
- starting pg2DM ch5
- final project written proposal due (1-2 pages) (50XP)
- presentation of final project proposal to class (50XP)
Thursday 7 October
- Home Practice #4: Do the three tasks on page 4-19 of pg2DM. Submit both the code and the answers to the questions to firstname.lastname@example.org (subject: 470: home practice #4)
- finish presentations
- pick proposal to work on and organize into teams
Tuesday 12 October
- Fall Break
Thursday 14 October
- FINAL EXAM questions start appearing
- Worksheet 5. Linear Regression
- Initial SCRUM meeting.
- Team Presentation: DM ch5
Tuesday 19 October
- SCRUM meeting – 5min.
- finish team presentation of ch 5
- git video
- git lab
- A bit of a review
- Precision and Recall
- Linear Regression
Thursday 21 October
- combined quiz 3&4: DM ch4 (4.6 to end), ch 5 & Bayes Formula. Know the following:
- Bayes Formula – know the formula in the middle of p90 for P(h|e) and be able to use it!
- know the algorithms (the ones in boxes) given in 4.6 to the end
- when would you use linear regression? perceptions? winnow?
- what are their disadvantages?
- Euclidean distance
- Instance based learning
- clustering, incl. nearest neighbor, k means, kD trees and ball trees.
- know what recall and precision are as well as false positives and false negatives
- Know when you would use ROC Curves, Recall-Precision Curves, Lift Charts, Cross Validation
- What is the most realistic evaluation method for clustering? For other data mining methods?
- team many-eyes.com demo
Tuesday 26 October
- Team Presentation: DM ch6
- Getting started with github video
- Regression and Perceptron worksheet
- Time for SCUM team discussion