I-Hsien Ting

Learn More
Introduction: The strategies that people use to browse Websites are difficult to analyse and understand: quantitative data can lack information about what a user actually intends to do, while qualitative data tends to be localised and is impractical to gather for large samples. Method: This paper describes a novel approach that combines data from direct(More)
E-commerce website design is a highly complex and subjective task which is often very difficult to get right from the start. Once a design has been implemented, customer behavior usually needs to be monitored in order to correct bugs and inefficiencies. While there are tools available that record and visualize customer click stream data, the information(More)
When analyzing patterns in server side data, it becomes quickly apparent that some of the data originating from the client is lost, mainly due to the caching of web pages. Missing data is a very important issue when using server side data to analyze a user's browsing behavior, since the quality of the browsing patterns that can be identified depends on the(More)
Social networking is gaining enormous popularity in the past few years. However, the popularity may also bring unexpected consequences for users regarding safety and privacy concerns. To prevent privacy being breached and modeling a social network as a weighted graph, many effective anonymization techniques have been proposed. In this work, we consider the(More)
This paper describes a novel web usage mining approach to discover patterns in the navigation of websites known as Unexpected Browsing Behaviours (UBBs). By reviewing these UBBs, a website designer can choose to modify the design of their website or redesign the site completely. UBB mining is based on the Continuous Common Subsequence (CCS), a special(More)
Location and local service is one of the hottest bunches of applications in recent years, due to the proliferation of <i>Global Position System</i> (GPS) and mobile web search technology. Spatial queries retrieving neighboring <i>Point-Of-Interests</i> (POI) require actual user locations for services. However, exposing the physical location of querier to(More)
Clickstream can be a rich source of data for analysing user behaviour , but the volume of these logs makes it difficult to identify and categorise behavioural patterns. In this paper, we introduce the Automatic Pattern Discovery (APD) method, a technique for automated processing of Clickstream data to identify a user's browsing patterns. The paper also(More)