Laxmi Ahuja

Approaches for Web Spam Detection

Spam is a major threat to web security. The web of trust is being abused by the spammers through their ever evolving new tactics for their personal gains. In fact, there is a long chain of spammers who are running huge business campaigns under the web. Spam causes underutilization of search engine resources and creates dissatisfaction among web community. Web Security being a prime challenge for search engines has motivated the researchers in academia and industry to devise new techniques for web spam detection. In this paper we present a comprehensive survey of techniques for detection of web spam and discuss their applicability and performance in various scenarios where they outperformed the others. We have categorized web spam detection with the primary focus on the approaches used for spam detection. The paper also gives the possible directions for future work.

Detecting redirection spam using multilayer perceptron neural network

Quality information retrieval from Web is essential for every search engine. But the quality of information is being exploited by spammers who make heavy use of malicious redirections for the purpose of phishing, downloading malware or attaining high search engine ranking. Malicious redirections present the irrelevant content to search user, thereby affecting user satisfaction. It also leads to wastage of network bandwidth. In this paper, we propose a neural framework for detecting redirection spam. We incorporated the feed-forward multilayer perceptron network and used scaled conjugate gradient algorithm that is able to perform very fast classification of URLs leading to redirection spam. We investigated the network empirically to choose the number of hidden layers and observed that when network is trained with two hidden layers, it gives better accuracy. We validated our proposed approach against the dataset of 2383 URLs and were able to detect the spammed redirections with high accuracy. The results indicate that neural networks are very effective technique to model the redirection spam detection.

A fuzzy logic approach for detecting redirection spam

Redirection spam is a relatively newer technique whereby spammers redirect the search user to an unwanted webpage or download malware on the victims machine without his consent. Spammers are making use of chained redirections to hide their nefarious activities. Detecting such malicious redirections is of prime importance for maintaining web security. In this paper, we have identified the factors that assist in detecting redirection spam and propose a fuzzy logic-based model for redirection spam detection. We validated our model against a set of URLs and were able to detect the spammed redirections with high accuracy.

Agile approaches towards Global Software Engineering

Global Software Engineering (GSE) is emerging trend in todays industry. The Software development industry is investigating the use of Agile development methodologies in distributed environment due to its benefits of better communication and coordination, improved productivity and quality. Global Software Engineering is continually achieving momentum from past four years. In last decade, the exceptional progress has been determined. New strategy models that are the acquainted from time with time in order to stay pace with third-dimensional requests of the business. New programming framework advancement standards are discovering the spot in the professional that Agile approaches towards Software System Development, use principally based Development and component based Development. As point of all the system models is same, i.e., to urge quality item, scale back time of improvement, efficiency change and decrease in cost. Still, no single strategy model is finished in itself. Programming framework business is moving towards. Deft programming framework Development. Programming framework improvement of spry systems to have the gotten the attention of programming framework architects and analysts around the world. Examination venture is all things considered rare. Coordinated programming advancement framework regardless of its own oddity is a significant space of examination at interims framework designing of programming order.

A technological perspective for evaluating component based technologies

Component-based Development (CBD) is a buzzword in Software Industry. Almost every software development Organization is using this concept for faster, quality oriented and less costly development. There are several technologies available in industry to support component-based software development. Present work studies two most popular component technologies J2EE from Sun Microsystems and .NET from Microsoft for various features like architecture, reliability, complexity, security and others and performs a comparative analysis for these technologies.

Development of expert search engine for web environment

This paper proposes a new expert web search engine for Web Environment. It applies a knowledge engineering based technique for the development of this expert system. To understand the basic functioning of search engine, various Web Forums and Blogs have been considered. Our work develops Intelligent Agent and Interaction Agent based knowledge base of Search Engine. This knowledge based Search Engine model thus developed will be useful in knowledge management and knowledge reuse. At user level, it can be used for suggesting best search results to the user and at organizational level it can be used for drawing various conclusions for managing quality database for better application use.

Label Count Algorithm for Web Crawler-Label Count

Web crawler is a searching tool or a program that glance the World Wide Web in an automated style. Through GUI of the crawler, user can specify the URL and all the links related are retrieved and annexed to the crawl frontier, which is a tally to visit. The links are then checked and retrieved from the crawl frontier. The algorithms for crawling the Web are vital when it comes to select any page which meets the requirement of any user. The present paper analyzes the analysis on the Web crawler and its working. It proposes a new algorithm, named as label count algorithm by hybridization of existing algorithms. Algorithm labels the frequently visited site and selects the best searches depending on the highest occurrence of keywords present in a Web page.

Factors Affecting Crime Against Women Using Regression and K-Means Clustering Techniques

The basic meaning of crime against women is direct or indirect mental or physical torture or cruelty towards women. Crime against women is increasing every year and as per the research they have doubled over the past ten years, according to latest data released by the NCRB (National Crime Records Bureau). As many as 2.24 million approx. crimes were reported against women over the past decade. On an average 25 crime per hour against women are reported, at least a complaint every two minutes. To control crime, the eyes have to be set on the factors which are influencing the crime against women. For this consideration there are various factors affecting the crime against women. In this paper factors are identified for crime against women. The impact of the individual factor has been checked for the overall crime rate in Delhi on the basis of regression analysis using SPSS tool and thereafter K-means clustering technique has been applied to classify the respondents or cases into clusters on the basis of degree of crime rate for various factors influencing the crime against women.

Performance Evaluation of Neural Network Training Algorithms in Redirection Spam Detection

Redirection spam is a technique whereby a genuine search user is forced to pass through a series of redirections and finally land on a compromised Web site that may present an unwanted content or download malware on his machine. Such malicious redirections are a threat to Web security and must be detected. In this paper, we explore the Artificial Neural Network algorithms for modeling redirection spam detection by conducting the performance evaluation of the three most used training algorithms, namely scaled conjugate gradient (trainscg), Bayesian regularization (trainbr), and Levenberg–Marquardt (trainlm). Our results indicate that the network trained using Bayesian regularization outperformed the other two algorithms. To establish the success of our results, we have used two datasets comprising of 2200 URLs and 2000 URLs, respectively.

Efficiency Evaluation of Recommender Systems: Study of Existing Problems and Possible Extensions

Recommender systems can help people to choose the desired product from a choice of various products. But there are various issues with the existing recommender systems. Thus, new systems which can efficiently recommend the most appropriate item to users based on their preferences are in demand. As a step towards providing the users with such a system, we present a general overview of the recommender systems in this paper. This paper also proposes potential solutions to the different problems which are found in current recommendation methods. These extensions to the current recommendation systems can improve their capabilities and make them appropriate for a broader area of applications. These potential extensions include integration of contextual information into the recommendation method, an improvement in understanding of items and users, developing less intrusive recommendation approaches, utilization of multi-criteria ratings, and providing more flexible types of recommendations.