Database Support for Data Mining Applications: Discovering Knowledge with Inductive QueriesRosa Meo, Pier L. Lanzi, Mika Klemettinen Springer, 2004 M07 28 - 332 pages Data mining from traditional relational databases as well as from non-traditional ones such as semi-structured data, Web data, and scientific databases housing biological, linguistic, and sensor data has recently become a popular way of discovering hidden knowledge. This book on database support for data mining is developed to approaches exploiting the available database technology, declarative data mining, intelligent querying, and associated issues, such as optimization, indexing, query processing, languages, and constraints. Attention is also paid to the solution of data preprocessing problems, such as data cleaning, discretization, and sampling. The 16 reviewed full papers presented were carefully selected from various workshops and conferences to provide complete and competent coverage of the core issues. Some papers were developed within an EC funded project on discovering knowledge with inductive queries. |
From inside the book
Results 1-5 of 68
... subset of the desired collection. This is the typical case when the theories are computed from a sample of the data (see, e.g., [74]) or when a relaxed constraint is used. Another important case of approximation for extended theories is ...
... subset of Items. A transactional database r is a finite and non empty multiset r = {t1 ,t2 ,...,t n} of transactions. Typical examples of transactional databases concern basket data (transactions are sets of products that are bought by ...
... subsets of a free set S have a different frequency than S. Notice that free itemsets have been formalized independently as the co-called key patterns [5]. Furthermore, the concept of free itemset formalizes the concept of generator [65] ...
... subsets of Items and can contain positive and negative items. In other terms, we want to have a symmetrical impact for the presence or the absence of items in transactions [14]. It leads to extremely dense transactional databases, i.e. ...
... subsets of each frequent itemset. This second step is far less expensive than the first one because no access to the database is needed: only the collection of the frequent itemsets and their frequencies are needed. Furthermore, the ...
Contents
1 | |
24 | |
Declarative Data Mining Using SQL3 | 52 |
Towards a Logic Query Language for Data Mining | 76 |
A Data Mining Query Language for Knowledge Discovery in | 95 |
Towards Query Evaluation in Inductive Databases Using Version Spaces | 117 |
The GUHA Method Data Preprocessing and Mining | 135 |
Constraint Based Mining of First Order Sequences in SeqLog | 154 |
Frequent Itemset Discovery with SQL Using Universal Quantification | 194 |
Deducing Bounds on the Support of Itemsets | 214 |
ModelIndependent Bounding of the Supports of Boolean Formulae | 234 |
Condensed Representations for Sets of Mining Queries | 250 |
Arnaud Giacometti Dominique Laurent Cheikh Talibouya Diop | 270 |
Evgueni N Smirnov Ida G SprinkhuizenKuyper H Japp van den Herik | 289 |
Kimmo Hätönen Mika Klemettinen | 304 |
Artur Bykowski Thomas Daurel Nicolas Méger Christophe Rigotti | 324 |