User Navigation Patterns Research Articles

Improving Mobile Web Navigation Using N-Grams Prediction Models

In this article, we propose to use N-gram models for improving Web navigation for mo-bile users. N-gram models are built from Web server logs to learn navigation patterns of mobile users. They are used as prediction models in an existing algorithm which improves mobile Web navigation by recommending shortcuts. Our experiments on two real data sets show that N-gram models are as effective as other more complex models in improving mobile Web navigation.

Evaluating Variable-Length Markov Chain Models for Analysis of User Web Navigation Sessions

Markov models have been widely used to represent and analyze user Web navigation data. In previous work, we have proposed a method to dynamically extend the order of a Markov chain model and a complimentary method for assessing the predictive power of such a variable-length Markov chain. Herein, we review these two methods and propose a novel method for measuring the ability of a variable-length Markov model to summarize user Web navigation sessions up to a given length. Although the summarization ability of a model is important to enable the identification of user navigation patterns, the ability to make predictions is important in order to foresee the next link choice of a user after following a given trail so as, for example, to personalize a Web site. We present an extensive experimental evaluation providing strong evidence that prediction accuracy increases linearly with summarization ability

Combined mining of Web server logs and web contents for classifying user navigation patterns and predicting users’ future requests

We present a study of the automatic classification of web user navigation patterns and propose a novel approach to classifying user navigation patterns and predicting users’ future requests. The approach is based on the combined mining of Web server logs and the contents of the retrieved web pages. The textual content of web pages is captured through extraction of character N-grams, which are combined with Web server log files to derive user navigation profiles. The approach is implemented as an experimental system, and its performance is evaluated based on two tasks: classification and prediction. The system achieves the classification accuracy of nearly 70% and the prediction accuracy of about 65%, which is about 20% higher than the classification accuracy by mining Web server logs alone. This approach may be used to facilitate better web personalization and website organization.

Clickstreams, The Basis To Establish UserNavigation Patterns On Web Sites

Collecting and mining clickstream data from e-commerce sites has become increasingly important for marketing, advertising, and traffic analysis activities. Organizations are promoting many initiatives concerning user’s navigation pattern discovery, in order to implement better sites, more functional and close to customers’ needs. Basically, the main idea is to provide more quality of attendance in their sites, and, consequently, get more profitability. However, clickstream processing is not a simple task. The sequences of clicks are very difficult to handle using conventional techniques, essentially due to their diversity and nature. They include a lot of aspects that reveal the multidimensional perspective of web data. OLAP technology provides today the means and techniques to represent, store and analyse such kinds of multidimensional data. However, it does not offer discovery driven analysis to support traversal pattern identification processes on web sites. Mining traversal pattern techniques can be applied in conjunction with OLAP as an integrated alternative for understanding those particular sequences of clicks. In this paper we present an integrated OLAP and mining approach specially conceived for exploring user navigation patterns based on clickstreams. We also describe the multidimensional structure provided for modelling click sequences and the OLAP operations and mining techniques that can be pushed over data cubes to bring up navigation patterns.

Efficient data mining for web navigation patterns

The concept of preference is proposed on the analysis of the present algorithms for mining user navigation patterns. It is based on the following hypothesis: if there are many different selections to leave a page, those selections that occur more frequently and the next page is viewed longer reveal user interest and preference. Representing user navigation interest and intention accurately by comparing relatively access ratio and the average of relatively access ratio of viewing time and selective intention, preference can be used for mining user navigation pattern instead of confidence. The higher preference, the more prefer to choose the selection. According to the conception, we propose two efficient algorithms based on the concept, UAM and PNT, which are developed for mining user preferred navigation patterns. Considering the structure of Web site, UAM can get user access preferred path by the page–page transition probabilities statistics of all users behaviours. PNT looks far into the past to correctly discriminate the different behavioral modes of the different users. Experiments show accuracy and scalability of the algorithms. It is suitable for applications in E-business, such as to optimize Web site or to design personalized service.

Web page clustering using a self-organizing map of user navigation patterns

The continuous growth in the size and use of the Internet is creating difficulties in the search for information. A sophisticated method to organize the layout of the information and assist user navigation is therefore particularly important. In this paper, we evaluate the feasibility of using a self-organizing map (SOM) to mine web log data and provide a visual tool to assist user navigation. We have developed LOGSOM, a system that utilizes Kohonen's self-organizing map to organize web pages into a two-dimensional map. The organization of the web pages is based solely on the users' navigation behavior, rather than the content of the web pages. The resulting map not only provides a meaningful navigation tool (for web users) that is easily incorporated with web browsers, but also serves as a visual analysis tool for webmasters to better understand the characteristics and navigation behaviors of web users visiting their pages.

Zipf's Law for Web Surfers

One of the main activities of Web users, known as ‘surfing’, is to follow links. Lengthy navigation often leads to disorientation when users lose track of the context in which they are navigating and are unsure how to proceed in terms of the goal of their original query. Studying navigation patterns of Web users is thus important, since it can lead us to a better understanding of the problems users face when they are surfing. We derive Zipf's rank frequency law (i.e., an inverse power law) from an absorbing Markov chain model of surfers' behavior assuming that less probable navigation trails are, on average, longer than more probable ones. In our model the probability of a trail is interpreted as the relevance (or ‘value’) of the trail. We apply our model to two scenarios: in the first the probability of a user terminating the navigation session is independent of the number of links he has followed so far, and in the second the probability of a user terminating the navigation session increases by a constant each time the user follows a link. We analyze these scenarios using two sets of experimental data sets showing that, although the first scenario is only a rough approximation of surfers' behavior, the data is consistent with the second scenario and can thus provide an explanation of surfers' behavior.

Analyzing user interactions with hypermedia systems

The unprecedented freedom for users to control the scope and sequence of their interactions with hypermedia systems presents many challenges to those who design and study these systems in educational settings. Early efforts to develop hypermedia systems revealed that the node-link structure of such systems is both advantageous and problematic (Conklin, 1987). When users have the freedom to follow any of a multitude of link permutations, disorientation often results. Further, without appropriate training, novice users do not possess the strategies necessary for effective "browsing" of large hypermedia documents (Duffy & Knuth, 1991). It has also been noted that the purpose for using the system or the task that the users are engaged in can influence patterns of interaction with hypermedia systems (Nelson, 1991). Many designers, therefore, advocate that features such as visual maps, database search facilities and guided tours be included in hypermedia systems to alleviate some of these problems (e.g. Hammond, 1989; Laurel, 1990, 1991).With the emergence of hypermedia systems as a major architecture for educational and other information-oriented software comes the related problem of how to document and analyze user interactions with such systems for the purposes of research and evaluation. There are a variety of interface design strategies that impact on how a system performs and should be evaluated. Many hypermedia systems to date have employed a "browsing" interface, but alternative approaches are also emerging (Nelson & Palumbo, 1992). Regardless of the type of interface, many questions can be generated when studying the interactions of users with hypermedia systems. For example, how many users chose to follow a particular link, and why was one link chosen over another? How does the choice of one link affect choices of subsequent links? When are graphic images, animations and video segments accessed? What user tasks are appropriate for guiding interaction with the system? What kinds of strategies do users develop while working with hypermedia systems? These and many other questions need answers when designing, developing and evaluating hypermedia applications for education and other settings, and provide the focus for this short article.A wealth of user interaction data can be easily collected within many hypermedia development environments in order to study aspects of the interface, including the nature of user navigation patterns, the time spent at each node and the use of help and orienting facilities. The data can represent the paths a user follows through the system, and the choices made at each node in the system. The problem is that because of the nature of this data, traditional methods of analysis such as surveys or pretest-posttest designs, are not particularly effective for determining usability or comparing alternative interface designs. Researchers have had to develop new techniques for analyzing patterns of user interaction in order to evaluate the design and effectiveness of hypermedia systems (Misanchuk & Schwier, 1992). There is a need to categorize and compare groups of users in order to compare the effectiveness of alternative system features, as well as describing characteristics of interaction by individual users within the same system.