A Skyline-based Decision Boundary Estimation Method for Binominal Classification in Big Data

Abstract

One of the most common tasks nowadays in a Big Data environment is the need to classify a large amount of data. There are numerous classification models, designed to perform best in different environments and datasets and each one of them has its advantages and disadvantages. However, when dealing with Big Data, their performance is significantly degraded because they are not designed or even capable of handling such large datasets. The current approach is based on a novel proposal of exploiting the dynamics of Skyline queries to efficiently identify the decision boundary and classify Big Data. A comparison against the popular k-NN, SVM and Naïve-Bayes classification algorithms shows that the proposed method is faster than the k-NN and the SVM. The novelty of this method is based on the fact that only small number of computations are needed in order to make a prediction, while its full potential is revealed in very large datasets.