Abstract The rapid developments in information and communication technologies and wireless communication networks have led to the utilization of the smart sensors. In modern healthcare applications, the utilization of smart sensors brings health professionals and patients together for intelligent and automated regular monitoring the activity of elderly people. Smart body sensors and smartphones are increasingly used for personal healthcare monitoring and wellbeing. One of the key improvements of smart sensor technologies in the healthcare monitoring system is the wearable sensor technology. Integrating of smart wearable sensors in healthcare has led to the development of smart applications such as smart healthcare and intelligent healthcare monitoring systems. Today, health informatics represent an important area to improve healthcare efficiency by optimizing the acquisition, storage, and the retrieval of crucial patient’s health information. In this context, there is a lot of excitement about the progress of machine learning techniques which play a crucial role in recognizing human activity. In this chapter, an intelligent smart healthcare system is presented to deliver pervasive human activity recognition (HAR) in an automated manner by using machine learning techniques in order to model and recognize activities of daily living in an accurate and efficient manner. Moreover, we focus on the dataset containing body motion and vital sign recordings from volunteers of diverse profiles while carrying out different physical activities for HAR purpose. In addition, this chapter shows how these approaches are used in two different datasets, a mobile phone and wearable body sensors, for robust and precise HAR. This study has shown that the identification of human activity based on sensor data is very challenging, especially with the existence of a variety of machine learning methods. When it comes to machine learning methods, there is no one solution to all.