Table of Contents
Mining Usage Data: Integrating Usage, Content, and Structure
Today
Simplified Web Access Layout
Back to Web Usage Mining Process
What’s in a Typical Server Log?
Usage Data Preprocessing
Example Page View
Log Entries for Page View
Problems in Identifying User Sessions and Transactions
Heuristics for Identifying User Sessions
Session Inference Example
Inferring User Transactions from Sessions
Use of Structure and Content for Usage Preprocessing
Quantifying Content and Structure
Data Preparation Tasks for Mining Content Data
Basic Automatic Text Processing 1
Basic Automatic Text Processing 2
Document Vectors
Assigning Weights
Vector Space Similarity Measure combine tf x idf into a similarity measure
Computing a similarity score
Computing Similarity Scores
E-Commerce Events
Product-Oriented Events
E-Commerce vs. Usage Data
Pattern Discovery
Session Analysis
Static Aggregation (Reports)
Online Analytical Processing (OLAP)
Common Data Mining Tasks
Sequential Patterns
Filtering Patterns Based on “Interestingness”
Domain versus Mined Knowledge
Algorithms for Filtering Patterns
Aggregate Usage Profiles
Methodologies for the Discovery of Aggregate Profiles
Profile Aggregation Based on Clustering Transactions (PACT)
Hypergraph-Based Clustering
Profiles Based on Hypergraph Clusters of Pageviews
Aggregate Usage Profiles - Hypergraph
Aggregate Usage Profiles - PACT
Integrating Content and Usage
Discovery of Content Profiles
|
Author: Bamshad Mobasher
Email: mobasher@cs.depaul.edu
Home Page: http://maya.cs.depaul.edu/~mobasher/classes/cs589
Download presentation source
|