Scribd Logo
Upload

Welcome to Scribd!

  • 7 views

    2 DHS IEEE DM Bank

    Uploaded by

    rajeshkumar32it
    This document summarizes a study that used different data mining programs and classification algorithms to analyze a bank marketing dataset from the UCI Machine Learning Repository. The study created classification models using k-nearest neighbor, Naive Bayes, and C4.5 decision tree algorithms in R, Knime, RapidMiner, and WEKA. It evaluated the models' accuracy, precision, and f-measure at different training/test data splits. The goal was to predict customer acquisition and compare algorithm performance across different data mining programs.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    2 DHS IEEE DM Bank

    Uploaded by

    rajeshkumar32it
    7 views1 page
    This document summarizes a study that used different data mining programs and classification algorithms to analyze a bank marketing dataset from the UCI Machine Learning Repository. The study created classification models using k-nearest neighbor, Naive Bayes, and C4.5 decision tree algorithms in R, Knime, RapidMiner, and WEKA. It evaluated the models' accuracy, precision, and f-measure at different training/test data splits. The goal was to predict customer acquisition and compare algorithm performance across different data mining programs.

    Original Description:

    Original Title

    2_DHS_IEEE_DM_Bank

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document summarizes a study that used different data mining programs and classification algorithms to analyze a bank marketing dataset from the UCI Machine Learning Repository. The study created classification models using k-nearest neighbor, Naive Bayes, and C4.5 decision tree algorithms in R, Knime, RapidMiner, and WEKA. It evaluated the models' accuracy, precision, and f-measure at different training/test data splits. The goal was to predict customer acquisition and compare algorithm performance across different data mining programs.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    7 views1 page

    2 DHS IEEE DM Bank

    Uploaded by

    rajeshkumar32it
    This document summarizes a study that used different data mining programs and classification algorithms to analyze a bank marketing dataset from the UCI Machine Learning Repository. The study created classification models using k-nearest neighbor, Naive Bayes, and C4.5 decision tree algorithms in R, Knime, RapidMiner, and WEKA. It evaluated the models' accuracy, precision, and f-measure at different training/test data splits. The goal was to predict customer acquisition and compare algorithm performance across different data mining programs.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 1

    IEEE TRANSACTION ON ELECTRIC ELECTRONICS COMPUTER SCIENCE, BIOMEDICAL ENGINEERINGS YEAR :2018

    Classification Of A Bank Data Set On Various


    Data Mining Platforms
    Bir Banka Müşteri Verilerinin Farklı Veri
    Madenciliği Platformlarında Sınıflandırılması
    Muhammet Sinan Başarslan İrem Düzdar Argun
    Computer Programming, School of Advanced Vocational Industry Engineering Department, Engineering Faculty
    Studies Düzce University
    Doğuş University Düzce, Turkey
    Istanbul, Turkey [email protected]
    [email protected]

    Abstract— The process of extracting meaningful rules from marketing data set. Bach et al. [2] have identified customers
    big and complex data is called data mining. Data mining has an who responded positively to the campaigns by performing
    increasing popularity in every field today. Data units are customer segmentation with a variety of methods such as
    established in customer-oriented industries such as marketing, artificial neural networks. Sumathi and Sivanandam have used
    finance and telecommunication to work on the customer churn data mining methods of financial institutions and other
    and acquisition, in particular. Among the data mining methods, institutions to discover interrelationships between data [3].
    classification algorithms are used in studies conducted for Keramati et al. have used decision trees, artificial neural
    customer acquisition to predict the potential customers of the networks, k-nearest neighbors and support vector machines,
    company in question in the related industry. In this study, bank
    among the machine learning algorithms, to predict existing
    marketing data set in UCI Machine Learning Data Set was used
    by creating models with the same classification algorithms in
    customers who would prefer competing banks using a
    different data mining programs. Accuracy, precision and f- telecommunication company data located in Iran. They have
    measure criteria were used to test performances of the identified the algorithm that gives the best result by comparing
    classification models. When creating the classification models, the the algorithms used in the study [4].
    test and training data sets were randomly divided by the holdout
    method to evaluate the performance of the data set. The data set II. OVERVIEW
    was divided into training and test data sets with the 60-40%, 75-
    25% and 80-20% separation ratios. Data mining programs used This section addresses the data mining and classification
    for these processes are the R, Knime, RapidMiner and WEKA. algorithms and data mining programs used throughout the
    And, classification algorithms commonly used in these platforms study.
    are the k-nearest neighbor (k-nn), Naive Bayes, and C4.5
    decision tree. A. Data Mining
    Data mining is the process of extracting meaningful and
    Keywords—data mining; banking; customer acquisition; data structures information in the complex data sets. During this
    mining programs.
    procedure, data mining methods such as classification,
    clustering and association rules are used. Data mining methods
    I. INTRODUCTION are used to analyze, categorize, summarize and determine the
    Today, data mining is used in the solution of problems in relationships using different dimensions of data [5]. These
    many fields such as health, finance and education. Data mining methods are divided into two groups as predictive or
    studies are being carried out in the field of health for diagnosis descriptive methods [7].
    of the disease, in customer-oriented industries such as
    telecommunication, insurance and banking to work on B. Classification Algorithms Used in the Study
    customer churn and customer acquisition. In this research, a In this study, bank marketing data set in UCI Machine
    forecasting study was carried out to see whether the campaign Learning Data Set [1] was used. Models were created using
    of a bank results in new customer acquisition. Another purpose classification algorithms on this data set. Classification
    of this study is to see the results of the same classification algorithms used in the study are the k-nearest neighbor (k-nn),
    algorithms in different data mining programs. Obtained results Naive Bayes (NB), and C4.5 decision tree. The classification
    were shown in tables in the results section. algorithms used are addressed in this section.
    There are many classification algorithms continuously
    developed for various applications in the literature on bank

    978-1-5386-5135-3/18/$31.00 ©2018 IEEE

    You might also like