Efficient Recovery of Incomplete Sensed Data in IoT Using Linear Regression

Volume 6 Issue 2 June - August 2018

Research Paper

Efficient Recovery of Incomplete Sensed Data in IoT Using Linear Regression

Chandra Sekhar Sanaboina*, Sai Krishna Kumar Madeti**

*Assistant Professor, Department of Computer Science and Engineering, JNTUK, Kakinada, Andhra Pradesh,India

** M.Tech Student, JNTUK, Kakinada, Andhra Pradesh, India.

Sanaboina ,C.S., & Madeti ,S.K.K. (2018). Efficient Recovery of Incomplete Sensed Data in IoT Using Linear Regression. i-manager’s Journal on Computer Science,6(2), 7-15. https://doi.org/10.26634/jcom.6.2.14885

Abstract

Though Machine Learning is a discipline in software engineering, it contradicts from the usual computing styles. In usual computing style, algorithms are sets of expressly programmed instructions utilized by computers to compute or solve the problem. Machine learning algorithms rather take into consideration for computers to prepare on information sources and utilize measurable examination keeping in mind the end goal to yield esteems that fall inside a particular range. This paper addresses five most commonly used classification algorithms, such as Logistic Regression, Naïve Bayes, K-Nearest Neighbors, Decision Tree, and Support Vector Machine. It aims to recover the incomplete sensed data of an IoT environment and proves that Linear Regression is the best suited for data recovery.