DaRoN: A Technique for Detection and Removal of Noise in IoT Data by using Central Tendency

Abstract

               The Internet of Things (IoT) is a significant technology that offers well-organized and trustworthy solutions for the innovation of many domains. Agriculture is one of the most concerned fields in IoT, where IoT based solutions are used to automate the maintenance and monitoring process with least human intervention. Large scale IoT based agricultural environment generates a large amount of data every moment. The agro-production environment is complex and there are numerous discrepancies in the collected raw data that cannot be directly traced by analysis and mining. To handle these inconsistencies in IoT agricultural data, this paper proposes a technique called Detection and Removal of Noise(DaRoN). The proposed technique removes the null values, error values, repeated values, incomplete values, and irrelevant values using measures of central tendency. In addition, a comparative analysis was performed with existingnoise removal techniques and the performance is measured using the Support Vector Machine(SVM) classifier. In this proposed research work, noisy data is eliminated to enhance classification accuracy. The DaRoN technique will be useful for improving the quality of collected data in agricultural environment.