Pages: 86-93

Abstract

Cloud computing is a new technology for managing and providing services over the Internet. Cloud ‎computing is a large group of connected computers. These computers can be personal computers or ‎network servers. They can also be public or private. The prediction of content error is one of the ‎fundamental challenges in cloud computing. In this research, it was tried to increase the accuracy of ‎prediction error of content in cloud computing using the nearest neighbor algorithm. The simulation ‎results in MATLAB software showed that the method for the data set had a very high performance. The ‎results were compared with the performance of three perceptron neural networks based on radial ‎performance and nearest neighbor. Finally, the best combination among these systems resulted in the ‎nearest neighbor error detection with nearly 74% accuracy‎‎‎.