Internet-of-Things (IoT) promises to give services to the users through connecting physical things using Internet. The conventional context aware system collects the data from users and stores it in cloud server. But, accuracy of classifying collected data using existing method was poor to store the user data with lower space complexity and to respond the user needed services with minimal time. In order to solve the above drawbacks, an Adaptive Discriminant Quadratic Boosting Classification based Radix Hash Cloud Data Storage (ADQBC-RHCDS) Model is proposed. The ADQBC-RHCDS Model is designed for providing the context aware IoT services to the cloud users with minimal response time and space complexity. In ADQBC-RHCDS model, Internet-ofThings (IoT) afford the services to the end cloud users by connecting an entity (i.e., person, place, or object) with sensors through Internet. Context Aware IoT helps to monitor and gather the information from users. After collecting the information, it is forwarded to the cloud server. Followed by, the cloud server classifies the collected information by designing Adaptive Discriminant Quadratic Boosting Ensemble Classifier (ADQBEC). After that, the classified data gets stored in the Radix Hash Tree Based Secured Cloud Data Storage (RHT-SCDS) for easy data access. Radix Hash Tree is a search tree used to store a set of data. Whenever the cloud user needs to access the data (i.e. insert or delete data), user sends the request to the cloud server. Then, cloud server provides the required services to the cloud user with minimal response time. Experimental evaluation of ADQBC-RHCDS model is carried out on factors such as classification accuracy, space complexity, and response time. The experimental result shows that the ADQBC-RHCDS model is able to reduce the space complexity and response time of context aware IoT services to the cloud users when compared to state-of-the-art works.