WEBKDD 2002 - Mining Web Data for Discovering Usage Patterns and Profiles

1. Front Matter

   PDF

2. LumberJack: Intelligent Discovery and Analysis of Web User Traffic Composition

   • Ed H. Chi, Adam Rosien, Jeffrey Heer

   Pages 1-16

3. Mining eBay: Bidding Strategies and Shill Detection

   • Harshit S. Shah, Neeraj R. Joshi, Ashish Sureka, Peter R. Wurman

   Pages 17-34

4. Automatic Categorization of Web Pages and User Clustering with Mixtures of Hidden Markov Models

   • Alexander Ypma, Tom Heskes

   Pages 35-49

5. Web Usage Mining by Means of Multidimensional Sequence Alignment Methods

   • Birgit Hay, Geert Wets, Koen Vanhoof

   Pages 50-65

6. A Customizable Behavior Model for Temporal Prediction of Web User Sequences

   • Enrique Frías-Martínez, Vijay Karamcheti

   Pages 66-85

7. Coping with Sparsity in a Recommender System

   • André Bergholz

   Pages 86-99

8. On the Use of Constrained Associations for Web Log Mining

   • Hui Yang, Srinivasan Parthasarathy

   Pages 100-118

9. Mining WWW Access Sequence by Matrix Clustering

   • Shigeru Oyanagi, Kazuto Kubota, Akihiko Nakase

   Pages 119-136

10. Comparing Two Recommender Algorithms with the Help of Recommendations by Peers

    • Andreas Geyer-Schulz, Michael Hahsler

    Pages 137-158

11. The Impact of Site Structure and User Environment on Session Reconstruction in Web Usage Analysis

    • Bettina Berendt, Bamshad Mobasher, Miki Nakagawa, Myra Spiliopoulou

    Pages 159-179

12. Back Matter

    PDF

1 WorkshopTheme Data mining as a discipline aims to relate the analysis of large amounts of user data to shed light on key business questions. Web usage mining in particular, a relatively young discipline, investigates methodologies and techniques that - dress the unique challenges of discovering insights from Web usage data, aiming toevaluateWebusability,understandtheinterestsandexpectationsofusersand assess the e?ectiveness of content delivery. The maturing and expanding Web presents a key driving force in the rapid growth of electronic commerce and a new channel for content providers. Customized o?ers and content, made possible by discovered knowledge about the customer, are fundamental for the establi- ment of viable e-commerce solutions and sustained and e?ective content delivery in noncommercial domains. Rich Web logs provide companies with data about their online visitors and prospective customers, allowing microsegmentation and personalized interactions. While Web mining as a domain is several years old, the challenges that characterize data analysis in this area continue to be formidable. Though p- processing data routinely takes up a major part of the e?ort in data mining, Web usage data presents further challenges based on the di?culties of assigning data streams to unique users and tracking them over time. New innovations are required to reliably reconstruct sessions, to ascertain similarity and di?erences between sessions, and to be able to segment online users into relevant groups.

• Alignment
• Session
• WWW
• algorithms
• association rule mining
• clustering
• data mining
• knowledge discovery
• recommender system
• usage analysis
• usage pattern discovery
• user profiling
• web access sequence mining
• web data mining
• web traffic mining

• University of Alberta, Canada

  Osmar R. Zaïane

• University of Minnesota, Minneapolis, USA

  Jaideep Srivastava

• Faculty of Computer Science, Otto-von-Guericke-University Magdeburg, Germany

  Myra Spiliopoulou

• Data Miners Inc., Boston, USA

  Brij Masand

Policies and ethics