Predictive Mining on Loan Data using Rattle

Venkata Sreedhar Ventrapragada

Abstract


Data mining on large database has been a major concern in research community due to the difficulty in analyzing huge volumes of data. Data mining refers to extracting or mining knowledge from large amounts of data. This paper attempts to explain data mining techniques, particularly “classification”, to predict which loan applicants are “risky” and which are “safe” for loan data set. For this purpose we use Rattle—“R Analytical Tool To Learn Easily” which is a powerful platform for data mining. Some preprocessing techniques such as Data Cleaning, Relevance Analysis, Data Transformation and Data Reduction have to be applied to improve the accuracy and efficiency of the classification process i.e removing noise by filling missing values, performing correlation analysis for identifying redundancies, scaling data to a specified range and performing principal component analysis (PCA) for reducing the dimensionality of data set. The main goal is to classify the loan applicants by classification based on decision tree method.


Keywords: Classification, Rattle, preprocessing, PCA, decision tree method.


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v4i10.1903

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science