Re[2]: Proposed book on Data Mining

[david_hatter]

     Dear Dorothy
     
     Thank you for offer to help on the Data Mining book.
     
     I have composed some copy, which runs to just over 5o lines and 
is 
     appended below.  If you could include it in your newsletter, Id 
be 
     grateful.
     
     With regards,
     
     Dave 
     
     ---------------------------------------------------------
     D.J.Hatter
     Publisher, Computing and Information Systems
     McGraw-Hill Publishing Company, 
     Shoppenhangers Rd, Maidenhead, Berkshire, England SL6 2QL
     
     Email: dave_hatter@mcgraw-hill.com
     
     Phone: (In order of probability)
     (mobile) +44 374 478508
     (home)   +44 1277 362915
     (office) +44 1628 502583
     
     fax: +44 1628 770224
     
     website:  http://www.mcgraw-hill.co.uk
     

-------------------------------------------------------------------------------

***************************************************************************
 
** Your views are requested on a proposed new publication in Data 
Mining ** 
***************************************************************************
At McGraw-Hill we have a proposal from Sarab Anand of Ulster 
University on the 
subject which is being considering for publication.  The proposal, 
which is 
summarised below, has been reviewed and has received praise for its 
technical 
and academic fidelity; we now need to assess the interest in the book 
among the 
informed community.  What I would like to ask, therefore, is to ask 
whether you 
would be interested in the book, for your own use or as a text for 
students. A 
brief e-note indicating your view, together with any observation 
which occurs to
you would help me greatly. An indication of the extent to which the 
subject 
appears in advanced u/g and p/g courses would be particularly useful. 
In the 
event of there being support for its publication we would be pleased 
to make it 
available at a preferred price for members of this group.

Thank you very much for your help. It is our view at McGraw-Hill that 
the book 
promises to be a significant addition to the literature and your 
response will 
assist us in our decision on whether to publish. Please address your 
response to
me, dave_hatter@mcgraw-hill.com

1: Introduction; Anand, Buchner, Hughes
Overview of Data Mining technologies. What Data Mining is and why it 
is needed. 
PART I:   Data Pre-Processing
2: Dealing with Missing Data; Ken Totton, Gavin Meggs, Blaise Egan 
(BT) Most 
common attribute value to bayesian and statistical models.
3: Data Dimensionality Reduction; Ron Kohavi(Stanford), 
McClean,Scotney (Ulster)
Covers techniques  to reduce the dimensionality of the data. 
4: Noise Modelling; Ray Hickey (Ulster)
"How can a discovery algorithm cope with inaccurate data" 
PART II Discovery Methodologies; Machine Learning Based Techniques
5: Rule Induction / Information Theory: Padhraic Smyth (U of 
California, Irvine)
The use of Information Theoretic measures within rule discovery is 
studied. 
6: Conceptual Clustering; A Doug Talbert, Doug Fisher, Vandebilt U, 
Tennessee 
Discusses problems in present clustering techniques & presents novel 
solutions. 
7: Heuristic Techniques; V. Rayward-Smith (University of East Anglia) 
Techniques
such as Simulated Annealing, Genetic Algorithms & hybrid techniques. 
8: 
Connectionism and Data Mining; Liu, Setiono (National U of Singapore)
This chapter discusses techniques available for rule extraction. 
Uncertainty 
Based Techniques:
9: Rough Set Analysis; Ivo Duntsch (Ulster), Gunther Gediga 
(Onsabruck, Germany)
Basic concepts &  two techniques for obtaining a logic of rough sets
10: Bayesian Belief Networks and L-L Modelling; Shapcott, Bell,  Liu 
(Ulster) 
Basic concepts of l-l models for two variables & their 
generalisation. Database 
Support for Data Mining:
11: Database Support for Attribute Oriented Induction;J.Han (Simon 
Fraser U) 
Attribute Oriented Induction operations mapped onto database 
operations.
12: Discovery in Distributed and Heterogeneous 
Databases;Bell,Anand,Hua (Ulster)
Initial work on requirements for distributed database support for 
discovery.
13: Distributed Statistical Databases; McClean, Scotney  (Ulster) The 
structure 
of a micro/macro data model and relations is examined. PART III The 
Role of the 
Human:
14: Using Background Knowledge; A. Tuzhilin (New York University) 
Covers the 
role of domain knowledge within Data Mining. 
PART IV Knowledge Post-Processing
15: Knowledge Filtering; Friedrich Gebhardt (GMD Labs, Germany )
Covers both aspects of interestingness discussing its different 
facets and 
providing a survey of measures used to address each of these facets. 
Covers both
objective as well as subjective measures.
16: Knowledge Validation;  Ken Totton, Gavin Meggs, Blaise Egan BT 
Labs, England
A number of different approaches to knowledge validation are 
reviewed.