Improved web usage mining techniques through data mining, user modeling, clustering, and behavioral heuristic methodologies

As the web has grown and evolved, researchers have developed techniques to extract useful information or knowledge from web resources such as web server access logs, a collection of techniques known as web usage mining. Web usage mining includes several techniques capable of identifying, grouping, and classifying online user behavior. We have created a series of methodologies as well as an automated framework capable of providing contextually meaningful usage information with minimal guidance from evaluators. The techniques discovered through our research have either improved or extended those found in existing data mining, user modeling, and user characterization techniques, and our automated framework assists evaluators in forming connections among the various aspects comprising the users' online experience.