Link prediction and path analysis using Markov chains
Proceedings of the 9th international World Wide Web conference on Computer networks : the international journal of computer and telecommunications netowrking
Model-Based Clustering and Visualization of Navigation Patterns on a Web Site
Data Mining and Knowledge Discovery
Hi-index | 0.00 |
AfterTheInjury.org, hosted by an academic medical center and launched in 2009, is an empirically-grounded Web-based intervention program. It provides information and psycho-educational interventional content for parents of injured children. This study analyzes the user navigational patterns and compares with the design of the intervention program. The web pages of the AfterTheInjury web sites were classified into four categories, namely Learn, Rate, Act, and Admin. Using the Google Analytics tools, we had collected the traffic data from January 1 to December 31 of 2009. We employed the Markov chain model to conduct the analysis. It was found that most users came to AfterTheInjury to learn followed by act and rate. Most users learned by visiting to a Learn page after another Learn page. Users were also learning about injury and finding out how to cope with injury alternatively. There was relatively lower probability of rating. It was not clear whether it was because there were only 3 Rate pages. We did not observe the pattern of Learn → Rate → Act as it was designed in the web site. However, we could not draw any conclusion yet because this preliminary study was limited by the navigation path length of 4. The navigation path analysis is useful to understand the user behavior patterns and determine if the hyperlink structure or graphical user interface design need to be enhanced. When certain desired patterns are missing in the analysis, we can decide how to create additional hyperlinks from specific content on the web page or modify the graphical user interface to guide users to specific information.