Abstract

Mobile-waste is a growing problem in India.  One of the factors that mobile has shorter life is rapidly changing technology. Most of the people throw their unwanted mobiles into scrap. Such scraped mobiles are hazardous to living life and environment as it contains lead, mercury, cadmium which affects living life and environment. It becomes necessary to recycle the mobile waste. According to the directives of Government of India, mobile manufacturer should recycle their own mobile-products which create e-waste. The classifying scrap mobile devices is a challenging task. This paper proposes work focus on automatic classification of damaged mobile as per the brands. A comparative study of different classifier algorithms such as Decision tree, Naive Bayes’ theorem, and support vector machine, K nearest neighbor, Convolution Neural Network and inception model was carried out. Accurately classifying damaged mobiles is an essential task for the manufacturer. This paper compares the performance of six machine learning algorithms. The primary objective is to evaluate the performance in classifying data with respect to classification test accuracy, precision and recall.