Schedule

Week 1

25 August

• what is data mining
• course logistics
• a simple musical artist data set

 

27 August

• getting started with iPython notebooks:
  • installing iPython
  • RunningCode.ipynb
• collaborative filtering
• nearest neighbor algorithm
• lab1 worksheet

Week 2

1 September

• Pearson Correlation Coefficient
• athletes.xls
• team worksheet
• data.xls
• Discussion of Chapter 2
• lab 2 worksheet

3 September

• RAT 1: Chapters 1 & 2
• lab time: lab 2
• homework 1 – due Sunday at 11:59pm.

• In the news (will show a sampling of these):
• Amazon ships before you order
• IBM Watson
• Low energy Bluetooth
  • What is it?
  • Where are you?
  • Health care
• Imagine the Future of Aging
• The Birth of a Word
• Movie Lens cont’d

 

Week 3

8 September

• Item Based Filtering
  • Adjusted Cosine Similarity
  • Slope One
• Weighted Slope One Worksheet
• Class Music Ratings
• Notebook 3 – Slope One
• work on the wine dataset
  • zip file containing code to get started
• • Movie Lens revisited.

10 September

• Slope One Worksheet – solution
• finish up tasks

Week 4

15 September

• Chapter 4: Intro
• mostly a lab day

17 September

• Chapter 4. Normalization
• Titanic Task

Week 5

22 September

• RAT #2: Chapters 3 & 4
• Entropy

24 September

Week 6

29 September

• Titanic Walkthrough
• link raw code for notebook

1 October

• Still Strict No Talking on my part.
• Titanic lab
• Entropy and Decision Tree continued.
• Beta version of Entropy Notebook

Week 7

6 October

• still no talking
• New Piazza Policy
• Titanic – sinking Intensive Lab

8 October

• Brand Spanking New version of Decision Tree/ Entropy notebook (150xp possible)
  • xp * 2 if demoed in class on 8 October (> 300 xp possible)
  • checkpoint 11 xp doubled if demoed before 16 October
• Decision Trees
• Decision Tree Notebook continued
• Covering Rules
• dimensionality

Week 8

13 October – Break

15 October

• Finish Entropy / Decision Tree notebook

• Talk on Classifying text.

Week 9

20 October

• RAT 4: Chapter 5, entropy. decision trees, Chapter 6 through page 6-27.
• Introduction to Naive Bayes

22 October

• Lab Day

Week 10

27 October

• Bayes Classifier Python Notebook

29 October

• Naive Bayes and unstructured text
  • demo book code with 20 newsgroup dataset 25xp
  • demo with another dataset 75-125xp.
  • Twitter Sentiment Analysis dataset

 

Week 11

3 November

• short talks on the following:
  • lift charts
  • dimension reduction
  • measures of accuracy
• Bayes tasks for XP

5 November

• Final Exam Posted
• Information Retrieval
• Tokenization and Normalization
• term document incidence matrices
• inverted index (postings index)

Week 12

10 November

• Final Exam Part 2 Posted
• clustering
• dog breed Google Sheet
• dog distances sorted
• Team Task- hierarchical clustering
• k-means clustering

12 November

• k-means clustering
• Building Watson
• Watson After Jeopardy
• information extraction and named entity recognition

Week 13

17 November

• Final Project discussed.
• Named Entity Recognition cont’d

19 November

Week 14

24 November

• last day to sign up for presentations
• Final Exam – final version posted
• Peer XP
• Matrix Factorization
  • Python Notebook

26 November – Thanksgiving

Week 15

1 Dec

• Presentations – Section 1:
  • Read the Web – Brian Will
  • Neural Networks – Joshua Mwanda
  • Deep Learning — Anna Corley, Matt O’Brien
  • The Future of Coaching Sports – Chelsea Irizarry, Lindsey Green, Brittany Raze
  • Dota 2 – Sepehr Sobhani
• Presentations Section 2:

3 Dec – last day

• Final Project Presentations