International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 08 Issue: 12 | Dec 2021 www.irjet.net p-ISSN: 2395-0072 © 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 427 Machine Learning Prediction of Human Activity Recognition Archit Jain 1 , Aayush Ranjan 2 , Mr. Ajay Kaushik 3 , Ms. Shallu Bashambu 4 1 Student, Dept of Information Technology, Maharaja Agrasen Institute of Technology, Delhi, India 2 Student, Dept of Information Technology, Maharaja Agrasen Institute of Technology, Delhi, India 3 Assistant Professor, Dept of Information Technology, Maharaja Agrasen Institute of Technology, Delhi, India 4 Assistant Professor, Dept of Information Technology, Maharaja Agrasen Institute of Technology, Delhi, India ---------------------------------------------------------------------***---------------------------------------------------------------------- Abstract – Wearable computing is becoming more and more incorporated into our daily lives. Wearable gadgets have recently attracted a lot of attention and widespread acceptance as a result of their compact size and decent processing power capabilities. These wearable gadgets with sensors (e.g. accelerometer, gyroscope, etc.) are excellent choices for tracking users' daily activities (e.g. walking, jogging, sleeping, and so on). Human Activity Recognition (HAR) has the potential to aid in the development of assistive technologies for the elderly, chronically ill, and those with special needs. Activity recognition can be used to offer information on patients' daily activities in order to aid the development of e-health systems such as Ambient Assisted Living (AAL). Despite the fact that human activity detection has been an active field for the development of context-aware systems for more than a decade, there are still critical issues that, if addressed, would represent a dramatic shift in how people interact with smartphones. A broad architecture of the essential components of any HAR system is described, as well as a data-gathering architecture for HAR systems. Machine learning techniques and technologies were used by HAR systems to generate patterns to characterize, evaluate, and predict data. Because a human activity recognition system should return a label such as walking, sitting, running, sleeping, falling, and so on, most HAR systems are supervised. The goal of this research is to use multiple machine learning methods on the UCI Human Activity Recognition dataset. Bagging with classification trees, logistic regression, support vector machines, random forest, and generalized linear model are among the machine learning algorithms or models that are used; Key Words: Human Activity Recognition, Machine learning, Wearable gadgets, HAR, Logistic Regression 1. INTRODUCTION Machine Learning is “the study of computer algorithms that improve automatically through experience. Applications range from data mining programs that discover general rules in large data sets, to information filtering systems that automatically learn users' interests.” - IEEE Definition. The IT industry's new dark is artificial intelligence (AI) and machine learning (ML). From gaming consoles to the administration of vast volumes of data at work, AI can be found in a variety of locations. Computer scientists and engineers are working hard to program machines with intelligent behavior, allowing them to think and react in real- time. AI has moved from a research topic to a stage where it is being implemented in businesses. The science and engineering of constructing intelligent devices, particularly computer programs, is known as artificial intelligence (AI). Artificial Intellect is akin to the job of utilizing computers to study human intelligence, although AI does not have to be limited to physiologically observable ways. Simply expressed, AI's goal is to make computers/computer programs clever enough to mimic the behavior of the human mind. Knowledge engineering research is an important part of AI research. In order to function and behave like people, machines and programs require a lot of knowledge about the world. To perform knowledge engineering, AI needs to have access to attributes, categories, objects, and the relationships between them. AI imbues machines with common sense, problem-solving, and analytical thinking abilities, which is a challenging and time-consuming task. Developers continue to increase artificial intelligence's powers and potential, despite ongoing disputes about its safety. Artificial Intelligence has come a long way since it was first imagined in science fiction. It became a requirement. AI, which is commonly used for processing and analyzing large amounts of data, aids in the handling of tasks that are no longer possible to complete manually due to their increased volume and intensity. Wearable devices, such as smartwatches, Google glasses, fitness trackers, sports watches, smart clothing, smart jewelry, implantable devices, and others, have recently attracted a lot of attention and widespread acceptance due to their small size, reasonable computation power, and practical power capabilities. These wearable gadgets, which are equipped with sensors (such as an accelerometer and a gyroscope), are a strong choice for tracking users' daily activities (such as walking, jogging, and smoking). Wearable and non-intrusive technologies for health and activity monitoring have become more common as wearable technology has advanced. The users are motivated to maintain a healthy lifestyle as a result of the continual monitoring of their lives and everyday activities. 2. HUMAN ACTIVITY RECOGNITION Human activity recognition has become a hot topic in the medical, military, and security fields. Patients with diabetes,