Discovery in Hydrating Plaster Using Multiple Machine Learning Methods

Published

August 3, 2003

Author(s)

John G. Hagedorn, J E. Devaney

Abstract

We apply multiple machine learning methods to obtain concise rules that are highly predictive of scienti cally meaningful classes in hydrating plaster over multiple time periods. We use three dimensional data obtained through X-ray microtomography at greater than one micron resolution per voxel at ve times in the hydration process: powder, after 4 hours, 7 hours, 15:5 hours, and after 6 days of hydration. Using statistics based on locality, we create vectors containing eight attributes for subsets of size 1003 of the data and use the autoclass unsupervised classi cation system to label the attribute vectors into three separate classes. Following this, we use the C5 decision tree software to separate the three classes into two parts: class 0 and 1, and class 0 and 2. We use our locally developed procedural genetic programming system, GPP, to create simple rules for these. The resulting collection of simple rules are tested on a separate 1003 subset of the plaster datasets that had been labeled with their autoclass predictions. The rules were found to have both high sensitivity and high positive predictive value. Using multiple machine learning methods we were able to go from unlabeled data to simple rules in a very straightforward manner.

Proceedings Title

Lecture Notes in Computer Science (LNCS) series

Volume

2534

Conference Dates

November 24-January 26, 2002

Conference Title

Fifth International Conference on Discovery Science, DS 2002, Lubeck, Germany.

Pub Type

Conferences

Citation

Hagedorn, J. and Devaney, J. (2003), Discovery in Hydrating Plaster Using Multiple Machine Learning Methods, Lecture Notes in Computer Science (LNCS) series, -1 (Accessed September 8, 2024)