Carbon emission measurement in improved cook stove using data mining

Abstract

Data mining is also known as knowledge discovery from huge data sets. Potential useful information also comes out from data set through data mining. This outcome is important for existing data sets and for further analysis, development and planning. This paper put a light on performance evaluation, based on the correct and incorrect instances of data classification using different classification algorithm. The paper sets out to make comparative evaluation of classifiers Naive Bayes, Multilayer Perception, J48 Decision Tree and IBK in the context of household datasets to maximize true positive rate and minimize false positive rate of defaulters rather than achieving only higher classification accuracy using WEKA tool. This paper also investigates and analyzes the existing raw data improved cook stoves (ICS) from different household information in Bangladesh; and also expedites Association Rule to extract some important information that can be helpful for future deployment, analysis and planning for sustaining efficient improve cook stove (ICS) program.