AAAI Publications, Twenty-Third International FLAIRS Conference

Font Size: 
Towards an Ensemble Framework for Assisting in Synthesis Tasks
Joseph Kendall-Morwick

Last modified: 2010-05-06


Ensemble methods have enjoyed much recent attention from machine learning researchers. Classification systems have been studied extensively through ensemble methods, and though efforts have also been made to study clustering and regression ensembles, some have suggested wider application of ensemble methods. In this paper, we seek to extend the application of ensemble methods to synthesis tasks involving the design of structures such as plans and workflows. To simplify the design task, incremental refinement recommendations are often sought involving explicit aspects of an incomplete or incorrect structure. We simplify such recommendations to two core components, the problem and the solution, in order to reduce the problem of generating recommendations into two analytical tasks. We formalize this distinction and explore the the implications of implementing an ensemble framework aware of both components, identifying additional requirements such a framework should satisfy to provide accurate and helpful recommendations.


ensemble methods; recommendation systems; machine learning

Full Text: PDF