Missing Data Imputation in Large Data Set Using Chernaïve Classifier Influenced by Bolzano–Weierstrass Theorem

Abstract

ABSTRACT
Data mining is a part of a super ordinate process called knowledge discovery in databases (KDD), where the term “database” refers to any kind of data storage and does not solely comprise data stored in database management systems. Data mining is integrated as a single step of the KDP, usually between model selection and interpretation. Knowledge Discovery in Databases (KDD) is the non-trivial process of identifying valid, novel, potentially useful, and ultimately understandable patterns in data. Knowledge discovery in databases describes the non-trivial process of identifying valid, novel, potentially useful, and ultimately understandable patterns in data. In other words, KDD addresses the problem of mapping low-level into other forms that might be more compact, more abstract or more useful. The key conjecture of the study highlights the transformation of the limitations of Multiple Imputation in Large Data set through Adaptive boosting Algorithm, Naïve Bayesian (NB), J48 algorithm, CHAID and to construct Chernaïve Classifier to enable Missing Data Prediction on defining the large data set. This research aims to prove that Chernaïve Classifier overcomes the limitations of Decision Tree (DT), Adaptive Boosting (ADAB), Naive Bayesian and J48. The influence of Bolzano–Weierstrass theorem on Missing data Imputation is the key objective of this article. A Mathematical model is constructed implementing the features of Chernaïve Classifier and treated over SONAR data set. This model overcomes the issue of independence classifier and boosting techniques to implement the prediction of missing data in the historical data items. To implement every stage of the research work, standard expertise tools like MATLAB, and SPSS for evaluation were used.

Keywords: Chernoff Bounds, Naive Bayes classifier, Decision Tree, Adaptive Boosting, J48, Missing Data Imputation, Classifier, Chernaïve Classifier, Bolzano–Weierstrass theorem