Gene Expression Programming Based Dataset Decoration for Improved Churn Prediction

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

@Misc{Trif:2013:IJERA,
  author =       "Silvia Trif and Adrian Visoiu",
  title =        "Gene Expression Programming Based Dataset Decoration
                 for Improved Churn Prediction",
  keywords =     "genetic algorithms, genetic programming, gene
                 expression programming, business intelligence, churn
                 prediction, classification, data mining",
  abstract =     "Mobile network operators rely on business intelligence
                 tools to derive valuable information regarding their
                 subscribers. A key objective is to reduce churn rate
                 among subscribers. The mobile operator needs to know in
                 advance which subscribers are at risk of becoming
                 churners. This problem is solved with classification
                 algorithms having as input data derived from the large
                 volumes of usage details recorded. For certain
                 categories of subscribers, available data is limited to
                 call details records. Using this primary data, a
                 dataset is created to be conveniently used by a
                 classification algorithm. Classification quality using
                 this initial dataset is improved by a proposed method
                 for dataset decoration. Additional attributes are
                 derived from the initial dataset through generation,
                 based on gene expression programming. Classification
                 results obtained using the decorated dataset show that
                 the derived attributes are relevant for the studied
                 problem.",
  annote =       "The Pennsylvania State University CiteSeerX Archives",
  bibsource =    "OAI-PMH server at citeseerx.ist.psu.edu",
  language =     "en",
  oai =          "oai:CiteSeerX.psu:10.1.1.418.6870",
  rights =       "Metadata may be used without restrictions as long as
                 the oai identifier remains attached to it.",
  URL =          "http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.418.6870",
  URL =          "http://www.ijera.com/papers/Vol3_issue3/GC3310851089.pdf",
}

Genetic Programming entries for Silvia Trif Adrian Visoiu

Citations