Web Browsing Behavior Research Articles

Smartphones vs. laptops

In this work we present the differences and similarities of the web browsing behavior in most common mobile platforms. We devise a novel Operating System (OS) fingerprinting methodology to distinguish different types of wireless devices (smartphone vs laptops) as well as operating system instances (iOS, Android, BlackBerry etc.). We showcase that most of the multimedia content in smartphone devices is delivered via Range-Requests, and a large portion of the video transfers are aborted. We also show that laptop devices have more intelligent browser caching capabilities. We investigate the impact of an additional browser cache, and demonstrate that a 10MB browser cache that is able to handle partial downloads in smartphones would be enough to handle the majority of the savings. Finally, we showcase that caching policies need to be amended to attain the maximum possible savings in proxy caches. Based on those optimizations the emulated proxy cache provides 10%-20% in bandwidth savings.

Prediction of User's Web-Browsing Behavior: Application of Markov Model

Web prediction is a classification problem in which we attempt to predict the next set of Web pages that a user may visit based on the knowledge of the previously visited pages. Predicting user's behavior while serving the Internet can be applied effectively in various critical applications. Such application has traditional tradeoffs between modeling complexity and prediction accuracy. In this paper, we analyze and study Markov model and all- Kth Markov model in Web prediction. We propose a new modified Markov model to alleviate the issue of scalability in the number of paths. In addition, we present a new two-tier prediction framework that creates an example classifier EC, based on the training examples and the generated classifiers. We show that such framework can improve the prediction time without compromising prediction accuracy. We have used standard benchmark data sets to analyze, compare, and demonstrate the effectiveness of our techniques using variations of Markov models and association rule mining. Our experiments show the effectiveness of our modified Markov model in reducing the number of paths without compromising accuracy. Additionally, the results support our analysis conclusions that accuracy improves with higher orders of all- Kth model.

Sequence-order-independent network profiling for detecting application layer DDoS attacks

Distributed denial of service (DDoS) attacks, which are a major threat on the Internet, have recently become more sophisticated as a result of their ability to exploit application-layer vulnerabilities. Most defense methods are designed for detecting DDoS attacks on IP and TCP layers and consequently have difficulty in detecting this new type of DDoS attack. With the profiling of web browsing behavior, the sequence order of web page requests can be used for detecting the application-layer DDoS (App-DDoS) attacks. However, the sequence order may be more harmful than helpful in the profiling of web browsing behaviors because it varies significantly for different individuals and different browsing behaviors. This article introduces a sequence-order-independent method for the profiling of network traffic and the detection of a new type of App-DDoS attacks. Four attributes are extracted from web page request sequences without consideration of the sequence order of requested pages. A model based on the multiple principal component analysis is proposed for the profiling of normal web browsing behaviors, and its reconstruction error is used as a criterion for detecting DDoS attacks. The proposed method is experimentally confirmed with various types of new App-DDoS attacks.

Design and implementation of a system for finding appropriate tags to photos in Flickr from Web browsing behaviour

In this paper, we illustrate the design and implementation of a system for finding appropriate tags to photos uploaded and shared in Flickr in detail. If a photo in Flickr can have good tags, it can be accessed by many users. However, it may be a difficult task to put many good tags to a photo. Our system discussed in this paper observes Web browsing behaviour of a user and attempts to make tag recommendation to photos automatically. Hierarchical clustering is exploited in our implementation. Some experimental results are reported for showing the effectiveness of our system.

Web user behavioral profiling for user identification

In this paper, we propose a simple, yet powerful approach to profile users' web browsing behavior for the purpose of user identification. The importance of being able to identify users can be significant given a wide variety of applications in electronic commerce, such as product recommendation, personalized advertising, etc. We create user profiles capturing the strength of users' behavioral patterns, which can be used to identify users. Our experiments indicate that these profiles can be more accurate at identifying users than decision trees when sufficient web activities are observed, and can achieve higher efficiency than Support Vector Machines. The comparisons demonstrate that profile-based methods for user identification provide a viable and simple alternative to this problem.

A Coherent Measurement of Web-Search Relevance

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> We present a metric for quantitatively assessing the quality of Web searches. The Relevance-of-Searching-on-Target index measures how relevant a search result is with respect to the searcher's interest and intention. The measurement is established on the basis of the cognitive characteristics of common user's online Web-browsing behavior and processes. We evaluated the accuracy of the index function with respect to a set of surveys conducted on several groups of our college students. While the index is primarily intended to be used to compare the Web-search results and tell which is more relevant, it can be extended to other applications. For example, it can be used to evaluate the techniques that people apply to improve the Web-search quality (including the quality of search engines), as well as other factors such as the expressiveness of search queries and the effectiveness of result-filtering processes. </para>

Extracting user preference from Web browsing behaviour for spam filtering

We focus on user behaviour that most e-mail users browse the web. In this paper, we attempt to exploit user preference extracted from the behaviour in a spam filtering method. The method reduces troublesome maintenance of the filter, since it keeps track of user preference as ham words in background. The ham words are used to determine whether a received e-mail is a ham or not. The method can detect some spams which are hard to classify correctly by existing Bayesian filters. We show that a combination of a Bayesian filter and our method reduces the number of false negatives.

Web browsing optimization over 2.5G and 3G: end‐to‐end mechanisms vs. usage of performance enhancing proxies

Abstract2.5 Generation (2.5G) and Third Generation (3G) cellular wireless networks allow mobile Internet access with bearers specifically designed for data communications. However, Internet protocols under‐utilize wireless wide area network (WWAN) link resources, mainly due to large round trip times (RTTs) and request‐‐reply protocol patterns. Web browsing is a popular service that suffers significant performance degradation over 2.5G and 3G. In this paper, we review and compare the two main approaches for improving web browsing performance over wireless links: (i) using adequate end‐to‐end parameters and mechanisms and (ii) interposing a performance enhancing proxy (PEP) between the wireless and wired parts. We conclude that PEPs are currently the only feasible way for significantly optimizing web browsing behavior over 2.5G and 3G. In addition, we evaluate the two main current commercial PEPs over live general packet radio service (GPRS) and universal mobile telecommunications system (UMTS) networks. The results show that PEPs can lead to near‐ideal web browsing performance in certain scenarios. Copyright © 2006 John Wiley &amp; Sons, Ltd.

The design and implementation of an agent-based framework for acceptable usage policy monitoring and enforcement

Reliance on the Internet in the workplace means that manually monitoring compliance with an Acceptable Usage Policy (AUP) is impractical given the volumes of data generated. Therefore, for such a system to function effectively, the processing of vast audit trails obtained must be processed by automated means. This paper introduces the incorporation of a novel user-monitoring framework into the domain of software agents for large-scale auditing of Internet use with possible extensions to general network use. It is intended that such an approach would replace current ad-hoc methods such as those based on perusing server logs with a more accurate representation of user activity. The system described herein is an experimental multi-agent one provisionally known as WebEngzilla, which actively monitors and reports on the Web browsing behaviour habits of network users unifying an ambient client monitoring system with a distributed data mining back end.

An empirical study of Web browsing behaviour: Towards an effective Website design

Prior studies have suggested that a good Website design which facilitates a user’s Web browsing behaviour would generally lead to better user performance. In this research, we examine user Website behaviour as a way to understand Website design using a “think aloud” protocol analysis. Main theoretical contributions of this research are the illustration of the flow of cognitive processes during the Website browsing and the establishment of Website design dimensions – the meaning and content implied by Website content, its outward form, and the structure and navigation – in relation to user performance.

Modeling individual differences using Dirichlet processes

We introduce a Bayesian framework for modeling individual differences, in which subjects are assumed to belong to one of a potentially infinite number of groups. In this model, the groups observed in any particular data set are not viewed as a fixed set that fully explains the variation between individuals, but rather as representatives of a latent, arbitrarily rich structure. As more people are seen, and more details about the individual differences are revealed, the number of inferred groups is allowed to grow. We use the Dirichlet process—a distribution widely used in nonparametric Bayesian statistics—to define a prior for the model, allowing us to learn flexible parameter distributions without overfitting the data, or requiring the complex computations typically required for determining the dimensionality of a model. As an initial demonstration of the approach, we present three applications that analyze the individual differences in category learning, choice of publication outlets, and web-browsing behavior.

Hotlists and Web browsing behavior—an empirical investigation

This paper constitutes an exploratory investigation into how individuals use the ‘bookmarks’ or ‘favorites’ feature of Web browsers, which has been generically designated a ‘hotlist.’ We conduct exploratory inquiries, first using interviews and then a Web-based survey of an Internet Service Provider's subscribers in order to examine Web users' construction and usage of hotlists as well as their ‘user orientation’ (goal-directed or experiential) toward the Web. Findings of both inquiries confirm the important role that is played by hotlists in browsing and searching behavior. Results also shed light on the determinants and consequences of hotlist construction and usage. In light of our findings, we discuss how user orientations may differ between more traditional media and the Web, draw managerial implications relating to the operation of pioneer brand advantages in the context of the Web, and offer suggestions for further research.

Psychological aspects of information seeking on the Internet

Two studies are presented that investigate information seeking behaviour on the Internet. In study one, soccer fans’ information seeking on the World Wide Web is investigated. In study two, access rates to a cancer information Web site are analysed. It is tentatively argued that there is a tendency for people to access information more commonly avoided in “real life”, although in the case of football fans, the tendency to “bask in reflected glory” remains when online, while cutting off reflected failure is minimised. Implications for understanding and researching psychological processes of Web browsing behaviour are discussed.

What is this page known for? Computing Web page reputations

The textual content of the Web enriched with the hyperlink structure surrounding it can be a useful source of information for querying and searching. This paper presents a search process where the input is the URL of a page, and the output is a ranked set of topics on which the page has a reputation. For example, if the input is www.gamelan.com, then a possible output is `Java'. We propose several algorithmic formulations of the notion of reputation using simple random walk models of Web-browsing behavior. We give preliminary test results on the effectiveness of these algorithms.