on Tuesday, June 23rd, 6:30 pm.

The Toronto APL Special Interest Group presents


A Panel Discussion on APL, J, OLAP and Data Mining

Panelists and Speakers:

1. Dan King
     "Dyadic Transpose: My Favourite Primitive for 
     Multi-dimensional Data Management" (short presentation)

2. Boyd Carter, DMR Consulting Group, Associate 
   Director of Data Warehousing
     "Overview of Data Warehousing"

3. Edwin M. Knorr
     "Outliers and Data Mining:  Finding Exceptions in 
     Large Datasets"

Follow-up: postscript version of Edwin's publication on 
this topic can be found at:
http://www.cs.ubc.ca:80/nest/dbsl/public/vldb98.ps


Also (a second short presentation):

1. Richard Levine
     "Delete Redundant Blanks: A Walk through APL Coding 
     Examples, from the Past to the Present" 

==============================================================

Abstracts:

1. "Overview of Data Warehousing"
     Boyd Carter, DMR Consulting Group, 
     Associate Director of Data Warehousing

Machiavelli
"There is nothing more difficult to take in hand, more 
perilous to conduct, or more uncertain in its success, than to 
take the lead in the introduction of a new order of things"

Machiavelli could have been instructing the young Prince in the 
building of a Data Warehouse.

Boyd Carter will identify one of the key streams of activity 
that can turn this negative view into a positive one. If you 
don't know where you are going, practically every path you take 
to get there will be wrong, especially when the path has as 
many branches as the Data Warehousing path.

He will describe how you can make a business case for Data 
Warehousing (know where you are going), track the probability 
of success through the construction of the Data Warehouse 
(take the right paths) and ensure that upon completion, the 
benefits described in the business case are realized (arrive at 
the planned destination). 

Boyd believes that:

"There  is  nothing  more challenging to take in hand, more 
exhilarating to conduct, or more rewarding in its success, than 
to take the lead in the introduction of a new order of things"
- A Data Warehouse.


2. "Outliers and Data Mining:  Finding Exceptions in 
   Large Datasets"
     Edwin M. Knorr

What do Wayne Gretzky, Chris Osgood, Alexander Mogilny, 
Ray Bourque, and Vladimir Konstantinov have in common? 
Answer: They're all outliers.

In some data mining applications, the patterns in the data are 
well established, but it is the exceptions to those patterns 
that are most interesting. The identification of outliers can 
lead to the discovery of truly unexpected knowledge in areas 
such as electronic commerce, credit card fraud, and even the 
analysis of performance statistics of professional athletes. 

In this talk, we study the notion of DB (Distance-Based)
outliers. Using examples, we show how to identify all 
distance-based outliers in large, multidimensional datasets. 

Our cell-based algorithm guarantees at most 3 passes over a 
dataset, and is appealing for disk-resident datasets of 
dimension k < 5. 
(Our nested loop algorithm is best for k >= 5 dimensions.)

This talk will describe outliers, data mining, existing 
statistical approaches for detecting outliers, an overview of 
our cell-based algorithm, performance results, and a case 
study involving NHL player statistics.

Ed Knorr is a PhD candidate in Computer Science at the 
University of British Columbia. His thesis is "Outliers and 
Data Mining:  Finding Exceptions in Large Datasets".
Ed also has an MSc degree from UBC and a BMath (co-op) degree 
from the University of Waterloo. He has over 10 years of 
experience as a systems programmer and database analyst in 
large, corporate environments. 

==============================================================
Date: TUESDAY, June 23, 1998
Time: 6:30 pm
Location: ROOM V-154 
(Television Studio "C" - ground floor, North side of the building, 
past elevator) Rogers Communication Building
80 Gould Street (North-East Corner of Church & Gould)
Toronto, Ontario
==============================================================

The Toronto APL Special Interest Group
http://www.torontoapl.org