Big data is a term used for defining the exponential growth and availability of both structured and unstructured and semi structured data is most difficult for processing with traditional database systems. Big data is inward from multiple sources at an alarming velocity, variety and volume. Although big data does not refer any specific quantity the term is used when speaking for pet bytes and Exabyte's of data. In existing method some keywords are used for indicating the user preferences and they used user based collaborative filtering algorithm for generating services to the users. They did not consider about the user preferences. In this paper we proposed a context aware approach for addressing the above challenges. The main aim of this paper is to make the users personalized recommendation for recommending the appropriate services to the users. Context aware collaborative filtering algorithm is used for meeting the requirements of users. I. INTRODUCTION Nowadays, the world may annotate with petabytes of electronic system also service details dispensed around various available global information databases. People of many methods as well as services are consistently distributing further details or giving the information databases with their choices and experience. Smartphone's as well as tablets operate as a window for monitoring and obtaining these annotated details and to render the consumers an input unit being accumulates further details. E-business, promoting and e-commerce is benefiting a lot by this persistent use of alternative device also customer data. Global industries, like eBay, Amazon, Apple iTunes or Google Play, provide millions of countless items as well as services in hundreds of types. These types cross a wide range of product groups from ordinary hardware to software also mobile applications. EBooks, gadgets, video and music streaming or even food. The large quantity of consistent obtainable details renders it complicated or even not possible. For consumers to manually choose an appropriate subset. As an individual user is not inclined to analyze all accessible data, the choice of that subset is of essential value for both, the