Improving MapReduce Based K-Means Algorithm Using Intelligent Technique

Data has expanded exponentially in recent years, leading to a search for ways of how to take advantage of these data, leading to the rise of big data science.The currently used data analysis techniques are not successful for different properties such as volume, velocity and variety, so, they are required to invent a new artificial intelligence technique.k-Means is one of the popular algorithms used to cluster large amounts of data and the problem with a k-Means is to pick the initial center for each unit which involves an optimization approach to solve the weak point such a Genetic algorithm which provides optimum performance in search algorithms that can be used to look for the initial point of each cluster rather than random selection of the initial centre.Big data involves many techniques, the most popular of which is the MapReduce which used to manage very large data.The proposed model is a Gk-Means based on the feature selection method parallel to MapReduce applying a Genetic algorithm to k-Means on selected data.