Machine and Deep Learning Techniques in IoT and Cloud

Abstract

Rapid advancements in communication technology have supported the invention of various internet-based devices. These devices communicate with one another and provide data from the physical world. Nowadays, the internet connected devices are used in various fields to make things easier. A great number of devices has been used, depending upon requirements. At the same time, the data produced by such devices is gradually increasing. To process the collected data, machine learning and deep learning techniques are applied. The Internet of Things (IoT) produces big datasets with multiple modalities but also a range of data with different quality standards. It is an important but also a challenging task to process all of the data within a certain time-frame. In this scenario, cloud computing gives us the optimal solution since the data generated is sent to distant cloud infrastructures. In addition to the cloud technology, machine learning (ML) and deep learning (DL) techniques are integrated with cloud computing to improve the effectiveness. In ML technique, the training data is given for learning to generate a set of rules from inferences on the data. Huge amounts of data that has been stored in the cloud gives input to DL techniques. DL architecture has been derived from the Artificial Neural Network (ANN) that uses multiple layers of nonlinear processing and transformation. The deep learning approach uses unknown elements in the input data to group objects, generate features and find new data patterns to build the model.