Structured induction is a method of applying machine learning in which a model for a task is learned using a representation where some of the components are themselves the outputs of learned models for specified sub-tasks. The idea was inspired by structured programming (Dahl, Dijkstra and Hoare, 1972), in which a complex task is solved by repeated decomposition into simpler sub-tasks that can be easily analyzed and implemented. The approach was first developed by Alen Shapiro (1987) in the context of constructing expert systems by decision tree learning, but in principle it could be applied using other learning methods.
Motivation and Background
Structured induction is designed to solve complex learning tasks for which it is difficult a priori to obtain a set of attributes or features in which it is possible to represent an accurate approximation of the target hypothesis reasonably concisely. In Shapiro’s approach, a hierarchy of decision treesis learned, where in each...
- Feigenbaum, E. A. (1977). The art of artificial intelligence: Themes and case studies of knowledge engineering. In R. Reddy (Ed.), Proceedings of the fifth international conference on artificial intelligence (IJCAI77) (pp. 1014–1029). Los Altos, CA: William Kaufmann.Google Scholar
- Gaines, B. (1996). Transforming rules and trees into comprehensible knowledge structures. In U. Fayyad, G. Piatetsky-Shapiro, P. Smyth, & R. Uthurusamy (Eds.), Advances in knowledge discovery and data mining (pp. 205–226). Cambridge, MA: MIT Press.Google Scholar
- Muggleton, S. (1987). Duce, an oracle-based approach to constructive induction. In IJCAI 87 (pp. 287–292). Los Altos, CA: Kaufmann.Google Scholar
- Quinlan, J. R. (1983). Learning efficient classification procedures and their application to chess end games. In R. Michalski, J. Carbonnel, & T. Mitchell (Eds.), Machine learning: An artificial intelligence approach, (pp. 464–482). Palo Alto, CA: Tioga.Google Scholar
- Razzak, M. A., Hassan, T., & Pettipher, R. (1984). Extran-7: A Fortran-based software package for building expert systems. In M. A. Bramer (Ed.), Research and development in expert systems (pp. 23–30). Cambridge: Cambridge University Press.Google Scholar
- Shapiro, A., & Niblett, T. (1982). Automatic induction of classification rules for a chess endgame. In M. R. B. Clarke (Ed.), Advances in computer chess (Vol. 3, pp. 73–91). Pergamon: Oxford.Google Scholar
- Shawe-Taylor, J., & Cristianini, N. (2004). Kernel methods for pattern analysis. Cambridge: Cambridge University Press.Google Scholar
- Zhang, J., & Honavar, V. (2003). Learning decision tree classifiers from attribute value taxonomies and partially specified data. In ICML-2003: Proceedings of the twentieth international conference on machine learning, Menlo Park, CA: AAAI Press.Google Scholar
- Zheng, Z. (1995). Constructing nominal X-of-N attributes. In Proceedings of the fourteenth International joint conference on artificial intelligence (IJCAI, 95) (pp. 1064–1070). Los Altos, CA: Morgan Kaufmann.Google Scholar