Credit scoring model: A combination of genetic programming and deep learning

Created by W.Langdon from gp-bibliography.bib Revision:1.3973

@InProceedings{Tran:2016:FTC,
  author =       "Khiem Tran and Thanh Duong and Quyen Ho",
  booktitle =    "2016 Future Technologies Conference (FTC)",
  title =        "Credit scoring model: A combination of genetic
                 programming and deep learning",
  year =         "2016",
  pages =        "145--149",
  abstract =     "In recent years, the market of customer lending grows
                 rapidly, that is a reason why credit scoring becomes a
                 core task of financial institutes. Many models based on
                 machine learning have been widely using and providing
                 robust performance. Because most machine learning based
                 models are black-box, it is hard to see the relations
                 between input data and scoring results. Therefore, this
                 paper focuses on improving both the accuracy and the
                 reliability of machine learning based model. Thus, we
                 propose a hybrid idea to combine the power of deep
                 learning network and the comprehensive genetic
                 programming which is extracted rules to build a robust
                 credit model. Our empirical experiment on
                 Australian/German customer credit data sets shows that
                 our model provides the best accuracy, highly reduce
                 credit risk, and reliable IF-THEN rules.",
  keywords =     "genetic algorithms, genetic programming",
  DOI =          "doi:10.1109/FTC.2016.7821603",
  month =        dec,
  notes =        "Also known as \cite{7821603}",
}

Genetic Programming entries for Khiem Tran Thanh Duong Quyen Ho

Citations