~ 62 ~ International Journal of Computing and Artificial Intelligence 2021; 2(1): 62-68 E-ISSN: 2707-658X P-ISSN: 2707-6571 IJCAI 2021; 2(1): 62-68 Received: 22-11-2020 Accepted: 25-12-2020 Kadiri Kamoru Oluwatoyin Department of Electrical and Electronic Engineering, Federal Polytechnic, Offa Kwara State, Nigeria Abubakar Dauda Department of Information and Communication Technology, National Control, Centre/Transmission Control Centre, Osogbo, Nigeria Enem Theophilus Aniemeka Department of Computer Science, Air Force Institute of Technology, Nigerian Air Force Base, Kaduna, Nigeria Corresponding Author: Kadiri Kamoru Oluwatoyin Department of Electrical and Electronic Engineering, Federal Polytechnic, Offa Kwara State, Nigeria Development of machine language for internet of things Kadiri Kamoru Oluwatoyin, Abubakar Dauda and Enem Theophilus Aniemeka DOI: https://doi.org/10.33545/27076571.2021.v2.i1a.28 Abstract The fast growth in the number of smart devices capable of running complex apps significantly impacts the information communication technology industry's landscape. The Internet of Things (IoT) continues to grow in popularity and relevance in man's daily existence. However, as the Internet of Things evolves, so do the associated problems. Thus, the need for IoT development and ongoing upgrading becomes stronger. To maximize the potential of IoT systems, machine learning technologies have recently been used. The implementation of machine learning algorithms in IoT systems is examined in detail in this paper. Two categories of machine learning-based IoT algorithms deal with fundamental IoT challenges like localization, clustering, routing, and data aggregation. Additional machine learning-based IoT algorithms deal with performance challenges like congestion control, fault detection, resource management, and security. Keywords: Wireless sensor networks (WSNs), machine learning, unsupervised learning, supervised learning, internet of things (IoT) 1. Introduction The Internet of Things (IoT) refers to a networking architecture that permits ubiquitous computing is pervasive and distributed services. The Internet of Things (IoT) is a network of interconnected gadgets and items that connect to the Internet and their surroundings. Everyday goods, such as sensors and smartphones/devices, may be linked to form a vast, interconnected system called the Internet of Things. About 50 billion IoT devices will be in use globally by 2020, producing more than 60 ZB data (Van der 2017; Sam 2016) [34, 32] . The Internet of Things will be made up of wireless sensor networks (WSNs) (IoT). Many people have been interested in WSNs in the last several years. The definition of a WSN includes a set of application-specific sensor nodes equipped with communication modules. Information gathered by the nodes is used to detect and record various environmental conditions. Notably, air temperature, humidity, wind speed, and wind direction are among the factors most often measured. The application-specific system development industry is highly suitable for WSNs. To work together and accomplish their jobs, sensor nodes in the IoT need a combination of WSNs and IoT. Both IoT and WSNs have a range of issues and concerns that must be solved. Energy efficiency, node placement, event schedule, route construction, data aggregation, defect detection, and data security are just a few applications for many systems. Due to the application of machine learning, this issue may be resolved. IoT performance and distribution will be significantly improved using machine learning. ML was first used as an approach to artificial intelligence in the 1960s (Ayodele 2010) [6] . Efforts to improve the resilience, effectiveness, and accuracy of algorithms have been continuous since that time. Machines that use machine learning algorithms to aid in a wide variety of applications, including bioinformatics, face and voice recognition, agricultural monitoring, fraud detection, and marketing, are often used today. Autonomous machine learning may be used to increase IoT systems' performance by analyzing previously gathered data and identifying the activities that led to better performance, and automating that process to perform better without the need to reprogram it. Machine learning plays a critical role in IoT applications because of the following: • IoT devices' monitoring of dynamic surroundings. Because of this, IoT systems that respond automatically to changes must be implemented. • Exploratory Internet of Things applications, like wastewater monitoring and volcanic