Managing & Mining Scientific Data

B669: Topics in Database and Information Systems
Spring 2000

Instructor: Ed Robertson and Dirk Van Gucht
Class: 17:30-20:00 Monday, Lindley 115
Newsgroup: ac.csci.b669

Seminar Goal & Objectives

The goal of this seminar is exposure to and involvement in database research at IU and elsewhere.
The primary focus is expected to relate to datamining and associated topics, but the key factor is research.

Objectives include:

Reading seminal and essential papers in the area
Projects related to data mining and database systems in general

Reading Topics

datacube and a priori algorithm
association rules
visual data mining
spatial databases
new query models -- metadata, text, XML, etc.

Possible Projects

visualization: mapping of multi-dimensional tabular data to 2-D presentations
formalization and implementation of a visual query interface incorporating metadata - MetaQBE
information theory based data mining
visualization: productizing magnification package
experimentation with datacube implementations
updating DocBase to include XML querying

Weekly Schedule

Date, Topic & Presenters	Readings
Jan. 10 Introduction, organizational session.
Jan. 17	Martin Luther King Day
Jan. 24 Spatial Databases Swoboda	Data Models and Query Languages for Spatial Databases, a general survey article.
Jan. 31 Data Mining Dalkilic	paper available in LH 401 lobby above the printer
Feb. 7 Information Dependencies Rossi	IU Tech Report 531 of that same title
Feb. 14 Visualization Groth
Feb. 21 QBE Vigo	Zloof paper
Feb. 28 Visualization Groth
Mar. 6 Martin
Mar. 13	Spring break
Mar. 20 Assoc. rules, apriori Groth	Fast Algorithms for MIning Association Rules, Agrawal & Shrikant
Mar. 27 Spatial databases Van Gucht
Apr. 3 Datacube Swoboda & Rossi	Datacube: A relational aggregation operator generalizing group-by, cross-tabs, and sub-totals, Gray et al.; Bottom-Up Computation of Sparse and Iceberg CUBEs, Beyer and Ramakrishnan, SIGMOD `99, also available via Beyer's web page
Apr. 10 Clustering&Classification
Apr. 17 final reports Swoboda & Rossi
Apr. 24 final reports Slominski, Vigo, & Martin

Links to Datamining Sites

Academic Programs
Data Miners at the University of Helsinki
Data Mining Institute, University of Wisconsin
DEVise: an Environment for Data Exploration and Visualization, University of Wisconsin
The Data Mine at U. Birmingham (UK), with many useful links including Other Sites Relevent To Data Mining
Machine Learning Repository at UCI
Industrial Sites
CWI (Centrum voor Wiskunde en Informatica) in Amsterdam: follow links for "Organization" and "Information Systems" (this site seems to generate URL's on the fly).
QUEST Data Mining Project at IBM Almaden
Mineset at SGI
TransBase HyperCube, which claims to be the first relational database system that fully integrates the UB-Tree, a multidimensional access method, into its kernel and query engine.
Student Projects and Sites
Rough Set tool from The Norwegian Univ. of Science and Technology
SCDS Create your own data sets, anyway you like, Simon Fraser Univ.

Class Enrollment and Authorization

The minimal background is B461, B561, or the equivalent.

Send email to edrbtsn@cs.indiana.edu for authorization. Be sure to include your student ID number. If you have not already taken a class with me, please include a synopsis of your database experience.

class roster

Managing & Mining Scientific Data

B669: Topics in Database and Information Systems
Spring 2000

Seminar Goal & Objectives

Reading Topics

Possible Projects

Weekly Schedule

Links to Datamining Sites

Class Enrollment and Authorization

PIZZA

Order pizza for this weeks class or place a standing order for pizza every week during class.

Last modified: Tue Jul 23 10:18:07 EST 2002

Managing & Mining Scientific Data

B669: Topics in Database and Information Systems Spring 2000

Seminar Goal & Objectives

Reading Topics

Possible Projects

Weekly Schedule

Links to Datamining Sites

Class Enrollment and Authorization

PIZZA

Order pizza for this weeks class or place a standing order for pizza every week during class. Last modified: Tue Jul 23 10:18:07 EST 2002

B669: Topics in Database and Information Systems
Spring 2000

Order pizza for this weeks class or place a standing order for pizza every week during class.

Last modified: Tue Jul 23 10:18:07 EST 2002