A Novel Ensemble Method For Secured Cloud Services For Storage And Performance Using Machine Learning Techniques

Abstract

Cloud computing render desired services to different users through the newline internet. The state-of-the-art context aware system aggregates the information from newline users and stores it in cloud server. Though, classification accuracy of collected data newline using conventional system was not higher to store the data on cloud server with newline minimal space complexity and to give needed services with lesser time. Besides, newline authentication is considered as significant problem to be addressed in order to newline increase the cloud security through verifying the identity of user. With the broad use newline of cloud storage, ensuring the integrity of user outsourced data is also gets greater newline concern to achieve higher data confidentiality. In order to resolve the above newline shortcomings in cloud computing, three efficient methods are developed as follows. newline An Adaptive Discriminant Quadratic Boosting Classification based Radix newline Hash Cloud Data Storage (ADQBC-RHCDS) Model is proposed to get better newline context aware cloud services performance via efficient classification and storage of newline cloud data. To increase the classification performance of data aggregated from user newline in cloud environment with minimal error rate and time, Adaptive Discriminant newline Quadratic Boosting Ensemble Classifier (ADQBC) algorithm is implemented in this newline work. Besides to that,Radix Hash Tree Based Secured Cloud Data Storage (RHT newlineSCDS)is designed in ADQBC-RHCDS Model to significantly stores classified user newline data on cloud server with minimal space and time complexity. As a consequence, newline proposed ADQBC-RHCDS Model decreases the amount of memory space utilized newline and response time taken for context aware cloud storage service in cloud. newline A Fast and Frugal Random Forest Decision Tree Classifier based Cloud User newline Authentication (FFRFDTC-CUA) method is proposed to get higher security newline performance during the cloud storage services rendering process via user newline verification. The fast and Frugal Random Forest Decision Tree Classifier newline (FFRFDTC) implemented in this proposed method to decreas