Discovery of functional relationships in multi-relational data using inductive logic programming

ILP systems have been largely applied to data mining classification tasks with a considerable success. The use of ILP systems in regression tasks has been far less successful. Current systems have very limited numerical reasoning capabilities, which limits the application of ILP to discovery of functional relationships of numeric nature. This paper proposes improvements in numerical reasoning capabilities of ILP systems for dealing with regression tasks. It proposes the use of statistical-based techniques like model validation and model selection to improve noise handling and it introduces a search stopping criterium based on the PAC method to evaluate learning performance. We have found these extensions essential to improve on results over machine learning and statistical-based algorithms used in the empirical evaluation study.

[1]  Ashwin Srinivasan,et al.  Numerical Reasoning with an ILP System Capable of Lazy Evaluation and Customised Search , 1999, J. Log. Program..

[2]  James Cussens.  Part-of-Speech Tagging Using Progol , 1997, ILP.

[3]  Sholom M. Weiss,et al.  Rule-based Machine Learning Methods for Functional Prediction , 1995, J. Artif. Intell. Res..

[4]  Saso Dzeroski,et al.  Declarative Bias in Equation Discovery , 1997, ICML.

[5]  Jianming Ye.  On Measuring and Correcting the Effects of Data Mining and Model Selection , 1998 .

[6]  Stephen Muggleton,et al.  Learning from Positive Data , 1996, Inductive Logic Programming Workshop.

[7]  Dick van Dijk,et al.  Forecasting industrial production with linear, nonlinear, and structural change models , 2003 .

[8]  David R. Anderson,et al.  Model Selection and Multimodel Inference , 2003 .

[9]  Stephen Muggleton,et al.  ILP for Mathematical Discovery , 2003, ILP.

[10]  Bongseog Jang,et al.  Threshold autoregressive models for VBR MPEG video traces , 1998, Proceedings. IEEE INFOCOM '98, the Conference on Computer Communications. Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies. Gateway to the 21st Century (Cat. No.98.

[11]  G. C. Tiao,et al.  Journal of the American Statistical Association Forecasting the U.s. Unemployment Rate Forecasting the U.s. Unemployment Rate , 2022 .

[12]  Aram Karalic,et al.  Employing Linear Regression in Regression Tree Leaves , 1992, ECAI.

[13]  J. Shao.  AN ASYMPTOTIC THEORY FOR LINEAR MODEL SELECTION , 1997 .

[14]  Richard A. Davis,et al.  Time Series: Theory and Methods , 2013 .

[15]  Ljup Co Todorovski.  Declarative Bias in Equation Discovery , 1997 .

[16]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[17]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.

[18]  Leslie G. Valiant,et al.  A theory of the learnable , 1984, STOC '84.

[19]  Ashwin Srinivasan,et al.  Compression, Significance, and Accuracy , 1992, ML.

[20]  H. Tong.  Non-linear time series. A dynamical system approach , 1990 .

[21]  Bernhard Sendhoff,et al.  Structure optimization of density estimation models applied to regression problems with dynamic noise , 1999, AISTATS.

[22]  B. Efron.  How Biased is the Apparent Error Rate of a Prediction Rule , 1986 .

[23]  Filip Zelezný,et al.  Learning Functions from Imperfect Positive Data , 2001, ILP.

[24]  Luc De Raedt,et al.  Top-Down Induction of Clustering Trees , 1998, ICML.

[25]  Rui Carlos Camacho de Sousa Ferreira da Silva,et al.  Inducing models of human control skills using machine learning algorithms , 2000 .

[26]  Kevin D. Hoover,et al.  Data mining reconsidered: encompassing and the general-to-specific approach to specification search , 1997 .

[27]  M J Sternberg,et al.  Structure-activity relationships derived by machine learning: the use of atoms and their bond connectivities to predict mutagenicity by inductive logic programming. , 1996, Proceedings of the National Academy of Sciences of the United States of America.