Globally speaking, stroke is a significant health problem. It has stayed the second-leading cause of death in the world since 2000. After the first two major causes of impairment, stroke is the third. Long-term disability has a substantial negative impact on people's capacity to lead productive lives. It is one of the major causes of extreme, ongoing weakness in all countries. Therefore, stroke represents a significant danger to global health. We use machine learning methods on the Healthcare Dataset Stroke to forecast strokes in this. This research utilizes the big data platform Apache Spark. Apache Spark, one of the most popular big data platforms, includes the MLlib library to handle enormous amounts of data. Apache Spark contains an MLlib library to manage huge data. MLlib, an API associated with Spark, offers machine learning methods. MLlib, an API associated with Spark, offers machine learning methods. A Multilayered Perceptron and Decision Trees are used to build the stroke prediction algorithm. These algorithms can help people and healthcare providers focus on health risks and changes in health status, which will eventually improve quality of life