World Wide Web Usage Mining

Slide 1: Usage mining for the World Wide Web (.pdf)

Slide 2: World Wide Web data mining

Slide 3: HTTP, URL, IP, domain name, and ICANN

Slide 4: World Wide Web usage mining

Slide 5: World Wide Web usage data gathering

Slide 6: Proxy-side logs

Slide 7: My prototypical WebQuilt

Slide 8: Web log information

Slide 9: Lynx (A World Wide Web text browser)

Slide 10: Using Lynx in C++ code to process a Web page

Slide 11: Using Lynx in Java code to process a Web page

Slide 12: Using java.net.* to process a Web page

Slide 13: World Wide Web usage data preparation

Slide 14: Building sessions

Slide 15: Navigation patterns discovery

Slide 16: Association rules discovery

Slide 17: Data clustering

Slide 18: Data clustering (cont.)

Slide 19: Data clustering (cont.)

Slide 20: Data classification

Slide 21: Data classification based on decision trees

Slide 22: WUM (A Web Utilization Miner)

Slide 23: WUM interfaces

Slide 24: The mining language MINT

Slide 25: The mining language MINT (cont.)

Slide 26: A WUM query example

Slide 27: Processing a mining query

Slide 28: A mining query processing example

Slide 29: MiDAS

Slide 30: Other ad hoc systems

Slide 31: Pattern analysis and visualization

Slide 32: Applications of Web usage mining results

Slide 33: Web caching

Slide 34: Major experimental/advanced Web usage mining systems and tools