The aim of this paper is to design a robust network security prediction model. The designed model is an intrusion detecting model constructed using neural networks. The intrusion detecting model detects anomaly and misuse-based attacks. Intrusion detecting model also performs three kinds of classification tasks. The tasks include classifying between occurrence of an attack or a normal case, classifying between different attack types or a normal case and classifying between occurrence of a attack or other case. The intrusion detecting model also shows the classification accuracy, execution time and amount of memory usage. The objectives of the intrusion detecting model are high accuracy, low execution time and minimum amount of memory usage. The intrusion detecting model constructed using neural networks, fulfils the objectives of high accuracy, low execution time and minimum amount of memory usage.

Many of the intelligent tutoring systems that have been developed during the last 20 years have proven to be quite successful, particularly in the domains of mathematics, science, and technology. They produce significant learning gains beyond classroom environments. They are capable of engaging most students' attention and interest for hours. This paper aims to establish some characteristics, properties and functions that an ITS should provide combined with speech, and the possible contributions that the different fields of research can make, proposing a multi-domain and multidisciplinary framework to address the research in this field. The framework incorporates a knowledge base where data and knowledge related to the problem are maintained and a model base related to student, teaching and environmental issues together with pedagogical perspectives. A theme underlying much of ITS research is domain independence, i.e. the degree to which knowledge encoded in the teaching model can be reused in different domains. Although to the external observer, domain independence seems like an essential characteristic of intelligence, many experts believe that some of the essential pedagogical knowledge in every domain is fundamentally domain-dependent. The proposed work was used for implementing ITS using supervised learning neural networks to a successful rate. Instead of being mere information-delivery systems, our systems help the students to actively construct knowledge.

No abstract

The neural network is one of the best data mining techniques that have been used by researchers in different areas for the past 10 years. Analysis on Indian stock market prediction using deep learning models plays a very important role in today's economy. In this chapter, various deep learning architectures such as multilayer perceptron, recurrent neural networks, long short -term memory, and convolutional neural network help to predict the stock market prediction. There are two different stock market price companies, namely National Stock Exchange and New York Stock Exchange, are used for analyzing the day-wise closing price used for comparing different techniques such as neural network, multilayer perceptron, and so on. Both the NSE and NYSE share their common details, and they are compared with various existing models. When compared with the previous existing models, neural networks obtain higher accuracy, and their experimental result is shown in betterment compared with existing techniques.