Random Forest Classifier Approach for Imbalanced Big Data Classification for Smart City Application Domains

Abstract

Information management and imbalanced classification is crucial for various domains namely medical diagnosis, text mining, tracking of financial transaction, telecommunication, and industrial and engineering applications for rapid development of smart cities. Upcoming needs of this digitized world comprise of utilization of the technologies which can handle complex unevenness within the data sample distribution within data. There are a variety of functional areas in smart cities which need to reshape unbalanced, complex and huge volume of data by incorporating classification techniques. Big data is moderately advance technology having potential to handle gigantic quantity of imbalanced data generating from real time applications like social networks etc. Random Forest Classification (RFC), comparatively outperforming classifier to handle disproportionate data distribution applications and gives utmost classification accuracy. Random Forest classification with big data (RFCB) deals with numerous technologies with parallel computing environment. RFCB are used to reduce the biasing of data towards majority and minority through sampling techniques. This paper incorporates the features of Big Data and extensive review of RFC on big data application for modeling the performance of classifier.