Computer Science and Engineering

In this paper, an integrated recommendation approach using Radial Basis Function Network (RBFN) and Collaborative Filtering (CF) is proposed. Radial basis function network is a neural network approximation method used to improve the accuracy of the recommendations. The proposed system RBFN_KFCM has offline and online phases. During offline, the system uses RBFN for smoothing and kernel fuzzy c-means (KFCM) method for clustering. During online recommendation, KFCM based approach is proposed. At the end of each session the clusters are updated by replacing the smoothed rating by the original rating. A comparison is made with benchmark recommender systems such as item-based, user-based systems, singular value decomposition (SVD) and also popular machine learning techniques such as Support Vector Machine (SVM), Multilayer Perceptron (MLP) using backpropagation algorithm, in terms of accuracy, decision-support measures and computational time. Empirical evaluation is carried out by using movielens dataset.

Recommender system a new marketing strategy plays an important role particularly in an electronic commerce environment. Among the various recommender systems, collaborative recommender system (CRS) is widely used in a number of different applications such as recommending web pages, movies, tapes and items. CRS suffers from scalability, sparsity, and cold start problems. An intelligent integrated recommendation approach using radial basis function network (RBFN) and collaborative filtering (CF), based on Cover's theorem, is proposed in order to overcome the traditional problems of CRS. The proposed system predicts the trend by considering both likes and dislikes of the active user. The empirical evaluation results reveal that the proposed approach is more effective than other existing approaches in terms of accuracy and relevance measure of recommendations.

Recommendations are the need of the hour in ecommerce sector. It allows user"s to shop according to their desire. Finding similarity between users to recommend items is the common concept used in most of the collaborative recommendation system. However there are many problems like sparsity, recommending items to cold-start user and handling copy-profile attack(Shilling"s attack).To overcome these setbacks we make use of trust and distrust statements and propose a novel method to increase the accuracy of recommended items.

Cloud computing is based on the concepts of distributed computing, grid computing, utility computing and virtualization. It is a virtual pool of resources which are provided to users via Internet. It gives users virtually unlimited pay-per-use computing resources without the burden of managing the underlying infrastructure. Cloud computing service providers' one of the goals is to use the resources efficiently and gain maximum profit. This leads to task scheduling as a core and challenging issue in cloud computing. This paper gives different scheduling strategies and algorithms in cloud computing.

With the advent of computing and communication technologies, it has become possible for a learner to expand his or her knowledge irrespective of the place and time. Web-based learning promotes active and independent learning. Large scale e-learning platforms revolutionized the concept of studying and it also paved the way for innovative and effective teaching-learning process. This digital learning improves the quality of teaching and also promotes educational equity. However, the challenges in e-learning platforms include dissimilarities in learner's ability and needs, lack of student motivation towards learning activities and provision for adaptive learning environment. The quality of learning can be enhanced by analyzing the online learner's behavioral characteristics and their application of intelligent instructional strategy. It is not possible to identify the difficulties faced during the process through evaluation after the completion of e-learning course. It is thus essential for an e-learning system to include component offering adaptive control of learning and maintain user's interest level. In this research work, a framework is proposed to analyze the behavior of online learners and motivate the students towards the learning process accordingly so as to increase the rate of learner's objective attainment. Catering to the demands of e-learner, an intelligent model is presented in this study for e-learning system that apply supervised machine learning algorithm. An adaptive e-learning system suits every category of learner, improves the learner's performance and paves way for offering personalized learning experiences.

In the current scenario, there is a tremendous necessity for strong cryptographic techniques for securely transmitting and storing data or information. The wide requirement of security in various areas develop the interest in doing research for producing variety of cryptographic algorithms which can provide security at various level. The algorithms can be implemented either in software or in hardware. The idea for secure algorithms evolved in the mid of 1970's. One of the most robust and secure asymmetric algorithm was proposed by Rivest, Shamir and Adelmann (RSA) in 1977 and proved to become a defacto standard, in cooperated with a large basis of products and applications that are still in operation. There are lots of work has been done in analysing the algorithm. Modular exponentiation is the basic operation for RSA. It consumes lots of time and resources for large values. To speed up the computation a naive approach is used in the exponential calculation in RSA by utilizing the Euler's and Fermat's Theorem .The method can be used in all scenarios where modular exponentiation plays a role.

In this chapter, a new data conceal technique is anticipated for digital images. The method computes the interpolation error of the image by using histogram shifting method and difference expansion. With the expectation of embedding high payload and less distortion, the undisclosed data has embedded in the interpolating error. Additionally for hiding the data, reversible data hiding technique is used. The histogram deviation is used as evidence for resulting the data conceal in the stereo images. To our best knowledge, by extracting the statistical feature from the image subsample works as steganalysis scheme. To enhance the revealing rate precision the well known support vector machine acts as classifier. In addition to that the experimental results show that the proposed steganalysis method has enhanced the detection exactness of the stego images.

Steganalysis is the art of detecting the stego images from the clean image. In the past few years, steganographers are hiding the secret messages in the most significant areas of images such as edges, curves, and some other tricky undetectable areas of the image. The small distortion channel is framed that fruitfully detects the edges and textures of an image and also it visually similar that greatly serves as a bounding frame for embedding the hidden messages. To detect the stego images, formal steganalysis methods are available based on clean images. This proposed work introduces the curvelet transform with soft thresholding to denoise the images for exposing the stego content. Experimental results show that various denoising methods are compared with curvelet transforms, the performance measures are taken, and it clearly shows that curvelet denoising method is state of the art than the other denoising method.

Web personalization systems are designed to improve user experiences by offering tailor-made services like web recommendations, adaptive web sites, web-enabled learning platforms, personalized web search etc. based on user interests and preferences recorded at log files of web, proxy servers or browsers. Personalization has varied applications through fittest solution meeting the model complexity and prediction accuracy. In this paper, we propose a framework to discover meaningful user browsing pattern on the web and convert the identified pattern into knowledge involving the web site being analyzed. Supervised machine learning techniques are used for discovering patterns of potential interests. Both single-valued, Bayesian inference & two-valued representations (belief & disbelief pair) Dempster-Shafer is proposed for inference and comparative precision identification. Proposed model thus addresses identification of browsing pattern that are known or expected in prior as well new p...

Objectives: Computer security is all about the study of cyber attacks with a view to defend against them. Cryptography is considered to be a class of science by using the special art of transforming information in a protected way such that it can overcome the attacks. There is an immense requirement of strong cryptographic algorithms in order to withstand against the various attacks. Methods: The Kerchoff''s Principle states that the encryption and decryption algorithms are always available to anyone. The security of the cipher against any sort of attack should be depends only on the secrecy of the key. There comes the play of cryptanalysis. It is art of breaking the keys by identifying the vulnerabilities existing in the systems. This paper deals with the classical encryption schemes and their cryptanalysis. The cryptanalysis for various encryption schemes differs a lot. Various cryptanalysis like statistical analysis, frequency analysis, trial and error (brute force) are carried out in this work. Findings: The classical encryption schemes such as caesar cipher, shift cipher, vigenere cipher, affine cipher and hill cipher are discussed in the paper. A detailed analysis about the security of the above mentioned ciphers are explored. Among the ciphers it is identified that if the key varies for each plaintext to be encrypted provides added security. But the worst is the management of the huge key space. A modified algorithm is proposed which can provide a better security using simple computations. In this approach lots of keys are used but generated from a single key by using simple shift and EXOR operation. In the main stream only one key need to be exchanged between the communication entities and for that exchange we can make use of the public key cryptosystem. Application: Cryptography is considered to be an ineluctable field in era of communication. Cryptographic algorithms acts as an underpinning for lots of applications such as Anonymous Remailers, Digital Signatures, Secured Money transactions etc.

Nowadays, Cloud Computing is a compelling paradigm for all enterprises where different services such as the server, storage, and application are delivered through the internet to the organization's computer and devices. It serves computing needs for both enterprise and end-user, so it became ondemand services as all enterprises emerging to cloud computing technology. So the cloud platform is incorporated in recommender a system that is mainly used for sorting a massive amount of data to identify user's interest by rating or feedback and makes item search easier for the end-user. Therefore, the Cloud-based Recommender system applied in a variety of applications to increase sales and user satisfaction in the market. Despite advances in recommender, the system still issues like sparsity, capability, and accuracy are needed to be addressed. In this paper, the cloud-based approach addresses the serious sparse data and capability issue by constructing a trustworthy recommender system. The proposed technique increases the density of the similarity matrix and coverage for accurate prediction. At the simulation result, the accuracy of recommendation is analyzed using evaluation metrics shows that the framework is useful and better prediction achievement.

A statistics reveals that the number of people selling goods over the internet has increased by more than 100percent since 2006. Almost everyone depend on the internet for everything such as reading newspapers, magazines, books and for searching research papers, to buy latest models of all gadgets and also for entertainment like hearing songs, watching movies, and for food recipes. The internet has changed the way of living. The reason behind this is 73% time consuming and still finding exactly what we need from information available is a tedious. We expect someone to recommend the best from huge data that fulfill ones need, tastes, behavior, interest etc. The "Information Overload"term was first coined by Alvin Toffler in his book named "Future Shock" in 1970 which is one of major issue the internet facing today. To address this issue and provide users best recommendations a System is developed called Recommender System. Recommender System applies various Data Mining methodologies to recommend efficiently for all active users based on their interest, preferences and ratings given for previous items and even based on similar users. In this paper we also analyze various issues and evaluation metrics used to measure the performance of the Recommender System.

A statistics reveals that the number of people selling goods over the internet has increased by more than 100percent since 2006. Almost everyone depend on the internet for everything such as reading newspapers, magazines, books and for searching research papers, to buy latest models of all gadgets and also for entertainment like hearing songs, watching movies, and for food recipes. The internet has changed the way of living. The reason behind this is 73% time consuming and still finding exactly what we need from information available is a tedious. We expect someone to recommend the best from huge data that fulfill ones need, tastes, behavior, interest etc. The "Information Overload"term was first coined by Alvin Toffler in his book named "Future Shock" in 1970 which is one of major issue the internet facing today. To address this issue and provide users best recommendations a System is developed called Recommender System. Recommender System applies various Data Mining methodologies to recommend efficiently for all active users based on their interest, preferences and ratings given for previous items and even based on similar users. In this paper we also analyze various issues and evaluation metrics used to measure the performance of the Recommender System.

E-health is rapidly picking up the gear and providing lots of services. It facilitates novel solutions for the problems suffered by the aged people as well as the people affected with chronic diseases. Apart from the above services, it devotes much interest in personal fitness. The healthcare system makes use of wearable wireless sensors which are implanted in patients, allow the monitoring of the health status at all time. The data collected from the patients transmitted through the network to the clinic or the doctor for a detailed diagnosis. Since healthcare applications are dealing with highly sensitive data, strong cryptographic algorithms should be incorporated for providing confidentiality. However, due to the computational and energy limitations of wireless devices we are forced to implement light weight ciphers which are computationally feasible in achieving confidentiality. This paper proposes a variant of hill cipher having two key matrices to add extra security to health related data.

In recent years, steganalyzers are intelligently detecting the stego images with high detection rate using high dimensional cover representation. And so the steganographers are working towards this issue to protect the cover element dependency and to protect the detection of hiding secret messages. Any steganalysis algorithm may achieve its success in two ways: 1) extracting the most sensitive features to expose the footprints of message hiding; 2) designing or building an effective classifier engine to favorably detect the stego images through learning all the stego sensitive features. In this chapter, the authors improve the stego anomaly detection using the second approach. This chapter presents a comparative review of application of the machine learning tools for steganalysis problem and recommends the best classifier that produces a superior detection rate.

Over the past era, subgraph mining from a large collection of graph database is a crucial problem. Existing works on subgraph mining is based on the threshold value, which returns similar graphs for a query graph. However, the number of graphs in the answer set for the same threshold can vary in terms of structure and context. In addition, scalability is another big problem due to insufficient storage.Further, it can suffer from security issues. In a distributed environment, since more queries come from different users, it is possible for the attackers to access the graph. These three problems highly exist in the current subgraph mining. To address this downside, our proposed work introduces aBlockchain-based Triune Layered Architecture for authenticated query search in the large scale dynamic graphs. The two fold process is handled in BTLA-LSDG: graph indexing and authenticated query search (query processing), which are implemented in triune layers (Data Generation Layer, Data Storage Layer, and Service Layer). Initially, data owners are authenticated to blockchain using Four-Q-Curve algorithm. The graph index is constructed by data owners and merged graph index is constructed by service providers. Based on the uploaded graph index, hash index is constructed using SHA-3. The hash index is fed into blockchain and form Dendrimer -Fractal Index. On the other hand, data user submits query with authentication. For every authenticated query, the four fold process is handled. Firstly, Multi-Constraint based Belief Entropy function is used for feature sets computation for a given query. Then Dual Similarity based MapReduce helps inmapping and reducing the relevant subgraphs with the use of optimal feature sets. Thirdly, Recurrent Neural Network (RNN)is used for subgraph isomorphic testing. Finally, graph index refinement process is undertaken to improve the query results. For this, query error is computed for each user's query at the end of retrieval. With respect to query error, fuzzy logic is used to refine the index of graph dynamically. This experiment is implemented using Hadoop environment and the results show better efficiency in terms of Scalability, Security and Storage. Further, it is tested with Precision, Recall, F-measure, Accuracy, Error Rate, Query Response Time and Positive Results.

Steganography deal with hiding information science, which offers an ultimate security in defence, profitable usages, thus sending the imperceptible information, will not be bare or distinguished by others. The aim of this paper is to propose a novel steganographic method in JPEG images to highly enrich a data security by RSA algorithm and attains higher payload by modified quantisation table. The goals of this paper are to be recognised through: 1) modify the quantisation table of the JPEG-JSTEG tool, hiding secret message with its middle frequency to offer great embedding capacity; 2) for challenge, secure RSA algorithm is used to prevent data from extraction. A broad experimental evaluation compares the performance of our proposed work with existing JSTEG was conducted. This algorithm resulted in greater PSNR values and steganogram histogram is more similar. Experimental results reported that the proposed system is a state-of-the model, contributing abundant payload and beating the statistical revealing. Besides, our method has better in all the parameters than JPEG-JSTEG method.

Language is a unique phenomenon that distinguishes man from other animals. It is our primary method of communication with each other, yet very little is understood about how language is acquired when we are infants. A greater understanding in this area would have the ...