Discovery in Hydrating Plaster Using Machine Learning Methods

Created by W.Langdon from gp-bibliography.bib Revision:1.3872

@InProceedings{conf/dis/DevaneyH02,
  author =       "Judith Ellen Devaney and John G. Hagedorn",
  title =        "Discovery in Hydrating Plaster Using Machine Learning
                 Methods",
  booktitle =    "5th International Conference on Discovery Science, DS
                 2002",
  year =         "2002",
  editor =       "Steffen Lange and Ken Satoh and Carl H. Smith",
  publisher =    "Springer",
  series =       "Lecture Notes in Computer Science",
  volume =       "2534",
  pages =        "47--58",
  address =      "L{\"u}beck, Germany",
  month =        nov # " 24-26",
  keywords =     "genetic algorithms, genetic programming",
  isbn13 =       "3-540-00188-3",
  URL =          "http://math.nist.gov/mcsd/savg/papers/discov2002.pdf",
  URL =          "http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.138.2341",
  DOI =          "doi:10.1007/3-540-36182-0_7",
  bibsource =    "DBLP, http://dblp.uni-trier.de",
  bibsource =    "OAI-PMH server at citeseerx.ist.psu.edu",
  contributor =  "CiteSeerX",
  language =     "en",
  oai =          "oai:CiteSeerXPSU:10.1.1.138.2341",
  abstract =     "We apply multiple machine learning methods to obtain
                 concise rules that are highly predictive of
                 scientifically 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 five 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 1000 of
                 the data and use the autoclass unsupervised
                 classification 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 1000 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. The
                 classes accurately identify important structural
                 components in the hydrating plaster. Moreover, the
                 rules identify the center of the local distribution as
                 a critical factor in separating the classes.",
}

Genetic Programming entries for Judith E Devaney John G Hagedorn

Citations