Veer Sain Dixit

An Overview on Tools for Peer to Peer Network Simulation

is an attempt to model a system in order to study it scientifically. Simulations are the most popular tool for examine peer-to-peer (P2P) applications. The cost of implementation of simulated model is less than that of large-scale experiments and, if carefully constructed, the simulated model can be more realistic than any tractable mathematical model. Simulating P2P overlay networks is a common problem for researchers and developers because P2P systems can consist of million of nodes and dynamic in nature. So that simulation for such a large dynamic network is difficult due to technical constraints even on the most powerful machines. In this paper we focus on various available P2P simulators and summarized them against a proposed set of attributes such as scalability, architecture language and pros and cons of each simulator.

Cloud Computing: State of the Art and Security Issues

Cloud Computing is the fastest growing technology in the IT industry. It helps in providing services and resources with the help of internet. Resources are always provided by the Cloud Service Provider. Resources may be servers, storage, applications and networks. Keeping Resources in the cloud environment can be helpful in saving infrastructure cost and time for the user. Transferring the entire information of the enterprise onto the cloud contains lots of security issues and threats. This paper focuses on the various concepts related to cloud computing, its various business and service models and its entities along with several issues and challenges related to it.

A propound method for agent based dynamic load balancing algorithm for heterogeneous P2P systems

In peer to peer (P2P) systems agent based load balancing is one of the most important problem. P2P systems are characterized by decentralization, scalability and dynamicity, such that they can been seen as instances of complex adaptive systems (CAS). In this paper we present ant-based load balancing algorithm, which effectively balances loads of peers distributed among P2P systems with the help of autonomous agents called Ants. Ants search a pair of overloaded and underloaded nodes through wandering on network and transfer tasks from different overloaded nodes to different underloaded nodes simultaneously. It is assumed that time break that ants spend on searching a pair of overloaded, underloaded node and transfer of virtual servers between them is negligible. The algorithm developed increases response time of submitted jobs and decreases communication overhead by load transfer in terms of virtual servers between overloaded and underloaded nodes simultaneously.

Generation of web recommendations using implicit user feedback and normalised mutual information

The knowledge base of a traditional web recommender system is constructed from web logs, reflecting past user preferences which may change over time. In this paper, an algorithm, based on implicit user feedback on top N recommendations and normalised mutual information, is proposed for collaborative personalised web recommender system. The proposed algorithm updates the knowledge base taking into account the changing user preferences, in order to generate better recommendations in future. The proposed approach and collaborative personalised web recommender systems without feedback are compared. Significant improvements are observed in precision, recall and F1 measure for proposed approach.

Scalable online product recommendation engine based on implicit feature extraction domain

This article presents a scalable and optimized recommender system for e-commerce web sites to maintain a better customer relationship management and survive among its competitors. The proposed system analyses the clickstream data obtained from an ecommerce site and predicts the preference level of the customer for the products clicked but not purchased using efficient classifiers such as decision trees, artificial neural networks and extended trees. Collaborative filtering technique is used to recommend products in which similarity measures are used along with efficient rough set leader clustering algorithm which helps in making accurate and fast recommendations. To determine the effectiveness of the proposed approach, an experimental evaluation has been done which clearly depicts the better performance of the system as compared to conventional approaches.

A Proposed Framework for Group-Based Multi-Criteria Recommendations

The Multi Criteria Group Recommender System is gaining attention. In the proposed framework, the authors intend to find solutions to two problems. First, group members assign arbitrary ratings to multiple criteria of items; these do not reflect their honest opinions. The proposed framework selects a promising group for the target user on the basis of demographic attributes. In case more than one group claims to be the promising group, the proposed framework implicitly derives multi-criteria ratings of each group member on multiple item attributes. Second, as the group size increases, difficulty in aggregating the preferences of all the group members increases, resulting in errors in group recommendations. The framework generates group recommendations from the nearest neighbor set of group members (GNNSet), instead of generating group recommendations from all the members of the target group. Cascade TOPSIS is proposed, which selects the GNNSet from the identified group thereby retaining only those group members who are likely to provide good quality group recommendations. TOPSIS on expert group members and nonexpert group members are used to handle group decision making. The preferences of members of the GNNSet are used to generate recommendations for the target user. Using the Movie Lens dataset, significant improvement in recommendation quality is observed.

Weighted difference entropy based similarity measure at two levels in a recommendation framework

Presenting small chunks of interesting information to a target user from the large pool of information available on World Wide Web are the primary task of a recommender system. Memory based Collaborative Filtering generates recommendations based on preferences of those users whose past preferences are similar to the current preferences of the target user. These users collectively form the Nearest Neighbor Set (NN Set) of the target user. The better the selection of NN Set, the better is the generation of recommendations. In the proposed scheme, the user ratings (preferences) on the available items are divided into two levels. Level I consist of ratings on popular items and Level II consist of ratings on unpopular items. The proposed similarity measure, “TSimD(UX1,UX2)”, between two users is based on weighted difference entropy. Modified memory based Collaborative Filtering calculates the proposed similarity measure at both the levels to improve the selection of users in the NN Set of the target user. It selects those users in the NN Set who are more similar to the target user with respect to items at Level II as compared to similarity between them with respect to items at Level I. The results on Movie Lens dataset depicted that the proposed similarity measure had higher accuracy than the Weighted Difference Entropy based similarity measure which worked on the entire preferences of the users. The proposed similarity measure was also compared with other similarity measures (like Cosine, Pearson Correlation, Spearman and Rating Frequency based Similarity Measure) which were also obtained at two levels.

Weighted-Frequent Itemset Refinement Methodology (W-FIRM) of Usage Clusters

Due to information overload on the Internet a large number of systems have been developed for extracting user behavior. This paper presents mining of Frequent Itemsets and refinement of usage clusters for web based applications. Here a particular case is under consideration where sessions in a cluster are in abundance, consequently leading to a very large number of not-so interesting recommendations for the user. To solve such problems we intend to refine clusters on the basis of Weighted Frequent Itemsets that in turn help to generate improved quality refined clusters. In the proposed work, Frequent Itemsets are sets of web pages that occur in sessions more than a given threshold known as the minimum support. Motivation for adapting Frequent Itemsets for refinement is the demand of dimensionality reduction. Experimental results show that the cluster quality using the proposed approach is better than the existing approaches (DBS, 2011 and HITS, 2010). After getting refined clusters the same can be used for number of applications such as Web Personalization, improvement in Web Site Structure, Analysis of Users’ Online Behavior and the services of a Recommender System.

OCRG: A proposed recommender for mitigating new user problem

In this paper, we propose an Online Cold Recommendation Generator (OCGR) to find recommendations for new users. It is based on their demographic attributes taking into account positive and negative ratings of other users. On the bases of these ratings, the proposed generator finds attraction, repulsion and balanced inclination of new users towards the existing items in the knowledge base. The results show that recommendations which are generated by using balanced inclination approach are less prone to rejection as compared to those recommendations which are generated by using only the attraction of new users towards existing items.

Refinement of recommendations based on user preferences

Collaborative Filtering is one of the most researched techniques. It generates recommendations from similar taste users in a group. In this paper, Information Theoretic Techniques are used to propose an Online Recommendation Generator based on Collaborative Filtering. It initially generates preliminary recommendations based on positive and negative user preferences and further refines these preliminary recommendations based on opposite user preferences. Experiments are conducted using MovieLens Dataset and considerable improvement in accuracy is seen in the results.