A major obstacle to fully integrated deployment of many data mining algorithms is the assumption that data sits in a single table, even though most real-world databases have complex relational structures. We propose an integrated approach to statistical modeling from relational databases. We structure the search space based on "refinement graphs", which are widely used in inductive logic programming for learning logic descriptions. The use of statistics allows us to extend the search space to include richer set of features, including many which are not boolean. Search and model selection are integrated into a single process, allowing information criteria native to the statistical model, for example logistic regression, to make feature selection decisions in a step-wise manner. We present experimental results for the task of predicting where scientific papers will be published based on relational data taken from CiteSeer. Our approach results in classification accuracies superior to those achieved when using classical "flat" features. The resulting classifier can be used to recommend where to publish articles.
|
866
|
Inductive Logic Programming
– Muggleton
- 1991
|
|
848
|
Conditional random fields: Probabilistic models for segmenting and labeling sequence data
– Lafferty, McCallum, et al.
- 2001
|
|
506
|
Information Theory and an Extension of the Maximum Likelihood Principle
– Akaike
- 1973
|
|
415
|
Algorithmic Program Debugging
– Shapiro
- 1982
|
|
382
|
A note on inductive generalization
– Plotkin
- 1970
|
|
306
|
Learning probabilistic relational models
– Friedman, Getoor, et al.
- 1999
|
|
288
|
Applied Logistic Regression
– Hosmer, Lemeshow
- 1989
|
|
261
|
Logic programming and Databases
– Ceri, Gottlob, et al.
- 1990
|
|
254
|
Enhanced hypertext categorization using hyperlinks
– Chakrabarti, Dom, et al.
- 1998
|
|
211
|
Digital Libraries and Autonomous Citation Indexing
– Lawrence, Giles, et al.
- 1999
|
|
189
|
Estimating the dimension of a model
– Schwartz
- 1978
|
|
164
|
Discriminative probabilistic models for relational data
– Taskar, Abbeel, et al.
- 2002
|
|
107
|
The missing link - a probabilistic model of document content and hypertext connectivity
– Cohn, Hofmann
- 2001
|
|
102
|
Shih: A Comparison of Prediction Accuracy, Complexity, and Training Time of Thirty-Three Old and New Classification Algorithms. Machine Learning 40(3
– Lim, Loh, et al.
- 2000
|
|
95
|
Discovery of frequent datalog patterns
– Dehaspe, Toivonen
- 1999
|
|
68
|
Propositionalization Approaches to Relational Data Mining
– Kramer, Lavrač, et al.
- 2001
|
|
67
|
Raedt. Relational reinforcement learning
– Dzeroski, de
- 1998
|
|
66
|
Relational Data Mining
– Dzeroski, Lavrac
- 2001
|
|
63
|
Iterative classification in relational data
– Neville, Jensen
- 2000
|
|
60
|
Relational Markov models and their application to adaptive Web navigation
– Anderson, Domingos, et al.
|
|
58
|
Using Web Structure for Classifying and Describing Web Pages. WWW2002
– Glover, Lawrence, et al.
- 2002
|
|
55
|
Feature construction with inductive logic programming: a study of quantitative predictions of biological activity aided by structural attributes
– Sriniviasan, King
- 1999
|
|
53
|
Linkage and autocorrelation cause feature selection bias in relational learning
– Jensen, Neville
- 2002
|
|
50
|
Induction of logic programs: FOIL and related systems
– Quinlan, Cameron-Jones
- 1995
|
|
48
|
Towards combining inductive logic programming with Bayesian networks
– Kersting, Raedt
|
|
47
|
Exploiting structural information for text classi cation on the WWW
– Furnkranz
- 1999
|
|
47
|
A statistical learning method for logic programs with distribution semantics
– Sato
- 1995
|
|
43
|
Relational learning with statistical predicate invention: Better models for hypertext
– Craven, Slattery
|
|
39
|
Aggregation-based feature invention and relational concept classes
– Perlich, Provost
- 2003
|
|
33
|
Relational learning via propositional algorithms: An information extraction case study
– Roth, &Yih
- 2001
|
|
32
|
First order regression
– Karalič, Bratko
- 1997
|
|
31
|
An introduction to inductive logic programming
– D˘zeroski, Lavra˘c
- 2001
|
|
31
|
Statistical relational learning for link prediction
– Popescul, Ungar
- 2003
|
|
28
|
Maximum entropy modeling with clausal constraints
– Dehaspe
- 1997
|
|
24
|
Top-down induction of Logical Decision Trees
– Blockeel
- 1998
|
|
20
|
Propositionalisation and aggregates
– Knobbe, Haas, et al.
- 2001
|
|
18
|
Towards structural logistic regression: Combining relational and statistical learning
– Popescul, Ungar, et al.
- 2002
|
|
11
|
Random effects models for network data
– Hoff
- 2003
|
|
9
|
Inducing classification and regression trees in first order logic
– Kramer, Widmer
- 2001
|
|
8
|
A proposal for learning by ontological leaps
– Foster, Ungar
- 2002
|
|
8
|
Statistical models for relational data
– Getoor, Koller, et al.
- 2002
|
|
7
|
Cumulativity as inductive bias
– Blockeel, Dehaspe
- 2000
|
|
1
|
Top-down induction of logical decision trees
– Raedt
- 1998
|
|
1
|
How to upgrade propositional learners to first order logic: A case study
– Laer, Raedt
- 2001
|
|
