The main problem dispersed with in this paper is to find a novel method for the improvement in the reliability analysis of Computer Network. Reliability prediction are estimated during the life cycle of a computer network with the aim of estimating failure. In designing a variable size network, the serviceability, availability and reliability of the any network is a primary consideration. The reliability calculation in varying size network is a problem of NP-hard; it requires more calculation and effort with the amplifying no of nodes and links. Many different approaches have been taken for reliability and probability calculation for triumphant communication between any pair of computers. The paper presents a method for identifying n-terminal network reliability based on RNN technique. The method derived in this paper preceding inputs which increases the speed of computation. The approach works efficiently and overcome the difficulties of the previous approaches defined with neural network model and other reliability estimation techniques. It is proposed that the RNN model be used to replace the most time-consuming component of the system reliability evaluation approach. A variable-length sequence input can be handled by RNN. The main goal of this paper is to predict asperity of reliability which is highly correlated with performance of network in any unfavorable conditions.

This Data mining is a technique for extracting useful information from large amounts of data. In large databases, enormous patterns may be examined and evaluated utilizing statistics and artificial intelligence. Data mining can be used to anticipate future trends or uncover hidden patterns. Classification, clustering, association rules, regression, and outlier identification are examples of data mining techniques. The data mining technology is receiving a lot of traction in the healthcare industry. In the discipline of bioinformatics, several researchers are using data mining techniques. Bioinformatics is the science of storing, retrieving, organizing, interpreting, and exploiting data from biological sequences and molecules. A prediction is a statement regarding a future event based on the current condition. The major intend of this work is to predict the microarray cancer using machine learning (ML) algorithms. Different phases are comprised in the prediction of microarray cancer. This research makes the implementation of voting-based classification algorithm. The suggested algorithm assists in optimizing the performance up to 2% while predicting the microarray cancer.