CAS CS 565, Data Mining Fall 2010

Schedule

Mon-Wed 4-5:30 pm

Course Outline

Data mining is the analysis of (often large) observational datasets to find unsuspected relationships and to summarize the data in novel ways that are both understandable and useful to the data analyst (Hand, Mannila and Smyth: Principles of Data Mining)

The goal of this course is to provide an introduction to the main topics in data mining including: frequent-itemset mining, clustering, classification, link-analysis ranking, dimensionality reduction etc. The focus of the course will be on the algorithmic issues as well as applications of data mining to real-world problems. Students will be required to solve small written and programming assignments that will help them better understand the covered material.

Instructor

Evimaria Terzi, evimaria@cs.bu.edu
Office Hours: Monday 2:30 pm - 4:00 pm and Tuesday 10:30 am -12:00 or by appointment.
www.cs.bu.edu/~evimaria

Workload

  1. Three programming projects

  2. Three problem sets

  3. Two exams; one midterm and one final

Prerequisites

Working knowledge of programming and data structures (CS 112, or equivalent). Familiarity with basic algorithmic concepts, probability, statistics and linear algebra. Programming projects will require knowledge of C (or C), java or python.

Syllabus

Many of these slides have been borrowed by lectures, talks and tutorials prepared by: Aris Anagnostopoulos, Petros Drineas, Piotr Indyk, George Kollios, Heikki Mannila, Panayiotis Tsaparas, Jeffrey Ullman.

The lectures on classification are using the slides from the book ‘‘Introduction to Data Mining" by Tan, Steinbach, Kumar, available here.

Sept 8 What is datamining/ Introductory lecture .pptx,.pdf
Sept 13 Mining frequent itemsets and association rules .pptx,.pdf
Sept 15 Condensed representations of itemset collections .pptx,.pdf
Sept 20 Homework 1; due October 4th download
Sept 20 Basics on Clustering .pptx,.pdf
Sept 22 Partition-based clustering .pptx,.pdf
Sept 27 Project 1; due Oct 13 2010 download
Sept 27 Hierarchical clustering .pptx,.pdf
Sept 29 Clustering Aggregation and Co-clustering .pptx,.pdf
Oct 4 Other clustering-related problems .pptx,.pdf
Sept 20 Homework 2; due October 18th download
Oct 6 Dimensionality reduction .pptx,.pdf
Oct 12 kd-trees.pptx,.pdf
Oct 13 Midterm
Oct 18 LSH .pptx,.pdf
Oct 20 Decision Trees.pptx,.pdf
Oct 25 k-NN and Naive Bayes Classification .pptx,.pdf
Oct 25 Project 2; due November 8 download
Oct 27 Support Vector Machines .pptx,.pdf
Nov 1 Evaluation and Ensembles .pptx,.pdf
Nov 3 Link Analysis Ranking .pptx,.pdf
Nov 8 PageRank - Topic Sensitive PageRank.pptx,.pdf
Nov 8 Project 3download,
Nov 10 Rank aggregation and voting systems.pptx,.pdf
Nov 15 Min cuts and cuts on graphs.pptx,.pdf
Nov 17 CS Colloq by Gerome Miklay
Nov 22 Min Cuts and Cuts on Graphs
Nov 22 Homework 3; due December 6 download
Nov 29 Spectral Graph Partitioning
Dec 1 Privacy-Preserving data mining .pptx,.pdf
Dec 6, 8 Time series analysis .pptx,.pdf
Week starting Dec 13 Final Exam

Reading material

Mining frequent itemsets and association rules

Rakesh Agrawal and Ramakrishnan Srikant: Fast Algorithms for Mining Association Rules. In International Conference on Very Large Databases, VLDB 1994.

Rakesh Agrawal, Heikki Mannila, Ramakrishnan Srikant, Hannu Toivonen, and A. Inkeri Verkamo: Fast discovery of association rules. In Advances in Knowledge Discovery and Data Mining, 307 - 328. AAAI Press, 1996.

Roberto J. Bayardo Jr.: Efficiently Mining Long Patterns from Databases. In SIGMOD Conference 1998: 85-93

F. Afrati, A. Gionis, H. Mannila: Approximating a Collection of Frequent Sets. In International Conference on Knowledge Discovery and Data Mining (SIGKDD), 2004.

Clustering

A. Gionis, H. Mannila, P. Tsaparas: Clustering Aggregation. In ACM Transactions on Knowledge Discovery from Data (TKDD), 2006.

Jon Kleinberg: An impossibility theorem for clustering. In NIPS, 2002.

A. Anagnostopoulos, A. Dasgupta, R. Kumar: Approximation algorithms for co-clustering. In PODS, 2008.

Kai Puolamaki, Sami Hanhijarvi, Gemma C. Garriga: An approximation ratio for biclustering. In Information Processing Letters, 2008.

Rakesh Agrawal , Johannes Gehrke , Dimitrios Gunopulos , Prabhakar Raghavan: Automatic subspace clustering for high-dimensional data. In SIGMOD, 1998.

Dimensionality Reduction

Christos Faloutsos, King-Ip Lin: FastMap: a fast algorithm for indexing, data-mining and visualization of traditional and multimedia datasets. In SIGMOD 1995.

Dimitris Achlioptas: Database-friendly random projections: Johnson-Lindenstrauss with binary coins. In Journal of Comp. & Sys. Sci.,special issue of invited papers from PODS’01.

Ella Bingham, Heikki Mannila: Random projection in dimensionality reduction: Applications to image and text data. In KDD, 2001.

For MDS read: Programming Collective Intelligence: Building Smart Web 2.0 Applications, Chapter 3, pages 49-52.

For Petros Drineas’ tutorials on linear-algebra methods for data mining see here.

For Piotr Indyk’s tutorial on Nearest-Neighbor search in low and high dimensions see here and here.

Similarity and Nearest-Neighbor (NN) search

Taher H. Haveliwala, Aristides Gionis, Dan Klein, Piotr Indyk: Evaluating strategies for similarity search on the web. In WWW Conference 2002.

Aristides Gionis, Piotr Indyk, Rajeev Motwani:Similarity Search in High Dimensions via Hashing. In VLDB 1999.

Link Analysis Ranking (LAR)

J. Kleinberg. Authoritative sources in a hyperlinked environment. Proc. 9th ACM-SIAM Symposium on Discrete Algorithms, 1998.

S. Brin, R. Motwani, L. Page and T. Winograd: The PageRank Citation Ranking: Bringing Order to the Web. Technical Report, Stanford Digital Libraries, 1998.

A. Borodin, J. S. Rosenthal, G. O. Roberts, P. Tsaparas: Link Analysis Ranking: Algorithms, Theory and Experimets. ACM Transactions on Internet Technologies (TOIT), Vol 5, No 1, February 2005.

Cuts on Graphs

David Cheng, Ravi Kannan, Santosh Vempala and Grant Wang:A Divide-and- Merge methodology for Clustering In ACM SIGMOD/PODS, 2005.

Ravi Kannan and Santosh Vempala and Andrian Vetta: On Clusterings: Good, Bad and Spectral. In the Journal of the ACM, May 2004.

Time-series segmentation

E. Terzi and P. Tsaparas: Efficient algorithms for sequence segmentation. In Siam Data Mining Conference (SDM), 2006.

Textbooks

There is no required textbook for the course. Suggested textbooks are:

Grading Policy

Incompletes will not be given

Late Assignment Policy: There will be a penalty of 10% per day, up to three days late. After that no credit will be given.

Collaborations/Academic Honesty

All course participants must adhere to the CAS Academic Conduct Code. All instances of adacemic dishonesty will be reported to the academic conduct committee