Machine learning-based IoT system for COVID-19 epidemics

Abstract

The planet earth has been facing COVID-19 epidemic as a challenge in recent time. It is predictable that the world will be fighting the pandemic by taking precautions steps before an operative vaccine is found. The IoT produces huge data volumes, whether private or public, through the invention of IoT devices in the form of smart devices with an improved rate of IoT data generation. A lot of devices interact with each other in the IoT ecosystem through the cloud or servers. Various techniques have been presented in recent time, using data mining approach have proven help detect possible cases of coronaviruses. Therefore, this study uses machine learning technique (ABC and SVM) to predict COVID-19 for IoT data system. The system used two machine learning techniques which are Artificial Bee Colony algorithm with Support Vector Machine classifier on a San Francisco COVID-19 dataset. The system was evaluated using confusion matrix and had a 95% accuracy, 95% sensitivity, 95% specificity, 97% precision, 96% F1 score, 89% Matthews correlation coefficient for ABC-L-SVM and 97% accuracy, 95% sensitivity, 100% specificity, 100% precision, 97% F1 score, 93.1% Matthews correlation coefficient for ABC-Q-SVM. In conclusion, the system shows that the process of dimensionality reduction utilizing ABC feature extraction techniques can boost the classification production for SVM. It was observed that fetching relevant information from IoT systems before classification is relatively beneficial.

PulmoNet: a novel deep learning based pulmonary diseases detection model

“…It has shown its efficacy by solving problems ranging from natural language processing to image classification using different DL and ML models. It makes predictions and inferences by analyzing the large quantity of input data, performs intelligent tasks such as feature detection, pattern recognition, translation, and perceptron on the data, and then makes the relevant decision [ 11 – 13 ]. Because of the high performance and other proven efficient qualities of AI in image identification and classification tasks, many researchers have conducted studies linking AI-founded techniques to the classification and prediction of different pulmonary diseases using either CT or X-ray imageries [ 14 – 17 ].…”

Enhancing Skin Cancer Detection and Classification in Dermoscopic Images through Concatenated MobileNetV2 and Xception Models

“…Over the last several decades [ 28 ], advancements in ML procedures have enhanced cancer identification accuracy by 15 and 20 percent. Deep learning (DL) is one of the domains of artificial intelligence that is expanding at the fastest rate owing to the many fields in which it is used [ 29 , 30 , 31 , 32 , 33 ]. DL, and especially CNNs, driven by advanced computer algorithms and enormous datasets, has become one of the greatest effective and widespread ML methods in picture recognition and categorization [ 34 ].…”

Open data-set identifier for open innovation and knowledge management

“…On the other hand, nowadays, one of the most important resources for companies, public organisations, governments and society in general is data. Contemporary society is characterised by the massive generation of data because of the use of technology and the development of Big Data and Machine Learning (Del Vecchio et al , 2018; Lytras et al , 2020; Gupta et al , 2021; Arowolo et al , 2022). In this context, this interest in data as assets and a necessary element for knowledge management has given rise to a movement advocating for the opening up of data-sets to innovate and create value for society.…”