Image Classification with Deep Learning Algorithms Using Knime Data Analytics Platform

In our age, data is everywhere, but not all of this data is available as meaningful and useful information. In the globalizing world, every individual and every machine is a data source and generates data with an ever-increasing acceleration. Our mobile phones, security cameras, bank cards, smart land irrigation systems, used ID cards, water meters and many more generate data. Obtaining meaningful and useful information from these produced data constitutes one of the biggest working areas of our age. In recent years, data science, which has developed rapidly, benefits from artificial intelligence technology that makes life easier in many ways. Health, education, industry and tourism are among the areas benefiting from artificial intelligence. Life is easier thanks to machine learning and deep learning algorithms, which are sub-branches of artificial intelligence. During the Covid-19 pandemic that emerged in China and affected the whole world, many systems using artificial intelligence were also developed. Thanks to the developed systems, this challenging process is aimed to be overcome with the least damage and in the shortest time. In this process, there are measures that can be taken that have a positive effect on the process. One of the precautions that can be taken is to wear a mask. In this thesis, it has been tried to determine whether people are wearing masks or not by using Knime data analysis program by using approaches based on deep learning. Thanks to the system created by training masked and unmasked images of people, it has been tried to determine whether the new data entering the system from the test data is masked or unmasked. In this study, AlexNet and Lenet deep learning architectures were used.