Fingerprint Gender Classification using Univariate Decision Tree (J48)

Data mining is the process of analyzing data from a different category. This data provide information and data mining will extracts a new knowledge from it and a new useful information is created. Decision tree learning is a method commonly used in data mining. The decision tree is a model of decision that looklike as a tree-like graph with nodes, branches and leaves. Each internal node denotes a test on an attribute and each branch represents the outcome of the test. The leaf node which is the last node will holds a class label. Decision tree classifies the instance and helps in making a prediction of the data used. This study focused on a J48 algorithm for classifying a gender by using fingerprint features. There are four types of features in the fingerprint that is used in this study, which is Ridge Count (RC), Ridge Density (RD), Ridge Thickness to Valley Thickness Ratio (RTVTR) and White Lines Count (WLC). Different cases have been determined to be executed with the J48 algorithm and a comparison of the knowledge gain from each test is shown. All the result of this experiment is running using Weka and the result achieve 96.28% for the classification rate.

@article{Abdullah2016,
  title     = {Fingerprint Gender Classification using Univariate Decision Tree (J48)},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {7},
  number    = {9},
  year      = {2016},
  publisher = {The Science and Information Organization},
  author    = {S. F. Abdullah and A.F.N.A. Rahman and Z.A. Abas and W.H.M. Saad},
  doi       = {10.14569/IJACSA.2016.070931},
  url       = {https://doi.org/10.14569/IJACSA.2016.070931}
}