web mining research : a survey
DESCRIPTION
Web Mining Research : A Survey. Raymond Kosala and Hendrik Blockeel ACM SIGKDD , July 2000 Presented by Shan Huang, 4/24/2007. Outline. Introduction Web Mining Web Content Mining Web Structure Mining Web Usage Mining Conclusion & Exam Questions. Four Problems. - PowerPoint PPT PresentationTRANSCRIPT
WebWeb MiningMining ResearchResearch: AA SurveySurvey
Raymond Kosala and Hendrik BlockeelACM SIGKDD , July 2000
Presented by Shan Huang,4/24/2007
Outline
IntroductionWeb MiningWeb Content MiningWeb Structure MiningWeb Usage MiningConclusion & Exam Questions
Four Problems Finding relevant information
Low precision Unindexed information
Creating new knowledge out of available information on the web
Personalizing the information Catering to personal preference in content and
presentation Learning about the consumers
What does the customer want to do? Using web data to effectively market products and/or
services
Other Approaches
Web mining is not the only approach Database approach (DB) Information retrieval (IR) Natural language processing (NLP)
In-depth syntactic and semantic analysis Web document community
Standards, manually appended meta-information, maintained directories, etc
Direct vs Indirect Web Mining
Web mining techniques can be used to solve the information overload problems: Directly
Attack the problem with web mining techniques E.g. newsgroup agent classifies news as relevant
Indirectly Used as part of a bigger application that addresses
problems E.g. used to create index terms for a web search service
The Research
Converging research from: Database, information retrieval, and artificial intelligence (specifically NLP and machine learning)
Paper focuses on research from the machine learning point of view
Outline
IntroductionWeb MiningWeb MiningWeb Content MiningWeb Structure MiningWeb Usage MiningConclusion & Exam Questions
Web Mining: Definition
“Web mining refers to the overall process of discovering potentially useful and previously unknown information or knowledge from the Web data.” Can be viewed as four subtasks Not the same as Information Retrieval Not the same as Information Extraction
Web Mining: SubtasksResource finding Retrieving intended documents
Information selection/pre-processing Select and pre-process specific information from selected
documentsGeneralization Discover general patterns within and across web sites
Analysis Validation and/or interpretation of mined patterns
Web Mining: Not IR or IE
Information retrieval (IR) is the automatic retrieval of all relevant documents while at the same time retrieving as few of the non-relevant documents as possibleWeb document classification, which is a Web Mining task, could be part of an IR system (e.g. indexing for a search engine)
Web Mining: Not IR or IE
Information extraction (IE) aims to extract the relevant facts from given documents while IR aims to select the relevant documents IE systems for the general Web are not feasible Most focus on specific Web sites or content
Web Mining and Machine Learning
As a broad subfield of artificial intelligence, machine learning is concerned with the development of algorithms and techniques that allow computers to "learn". Web mining not the same as learning from the Web.Some applications of machine learning on the web are not Web MiningSome methods used for Web Mining besides machine learningHowever, there is a close relationship between web mining and machine learning.
Outline
IntroductionWeb MiningWeb Content MiningWeb Structure MiningWeb Usage MiningConclusion & Exam Questions
Web Mining CategoriesWeb Content Mining Discovering useful information from web contents/data/documents. IR view for finding DB view for modeling
Web Structure Mining Discovering the model underlying link structures (topology) on the
Web E.g. discovering authorities and hubs
Web Usage Mining Make sense of data generated by surfers Usage data from logs, user profiles, user sessions, cookies, user
queries, bookmarks, mouse clicks and scrolls, etc.
Web Content Data Structure
Unstructured – free textSemi-structured – HTMLMore structured – Table or Database generated HTML pagesMultimedia data – receive less attention than text or hypertext
Web Mining: The Agent Paradigm
User Interface Agents information retrieval agents, information filtering
agents, & personal assistant agents.
Distributed Agents distributed agents for knowledge discovery or data
mining. Problem solving by a group of agents
Mobile Agents
Web Mining: The Agent Paradigm
Content-based approach The system searches for items that match based
on an analysis of the content using the user preferences.
Collaborative approach The system tries to find users with similar
interests Recommendations given based on what similar
users did
Outline
IntroductionWeb MiningWeb Content MiningWeb Structure MiningWeb Usage MiningConclusion & Exam Questions
Web Content Mining: IR View
Unstructured Documents Bag of words, or phrase-based feature
representation Features can be boolean or frequency based Features can be reduced using different feature
selection techniques Word stemming, combining morphological
variations into one feature
Web Content Mining: IR View
Semi-Structured Documents Uses richer representations for features, based on
information from the document structure (typically HTML and hyperlinks)
Uses common data mining methods (whereas unstructured might use more text mining methods)
Web Content Mining: DB ViewTries to infer the structure of a Web site or transform a Web site to become a database Better information management Better querying on the Web
Can be achieved by: Finding the schema of Web documents Building a Web warehouse Building a Web knowledge base Building a virtual database
Web Content Mining: DB View
Mainly uses the Object Exchange Model (OEM) Represents semi-structured data (some structure,
no rigid schema) by a labeled graph Process typically starts with manual selection of Web sites for content miningMain application: building a structural summary of semi-structured data (schema extraction or discovery)
Outline
IntroductionWeb MiningWeb Content MiningWeb Structure MiningWeb Usage MiningConclusion & Exam Questions
Web Structure Mining
Interested in the structure between Web documents (not within a document)Inspired by the study of social networks and citation analysisExample: PageRank – GoogleApplication: Discovering micro-communities in the WebMeasuring the “completeness” of a Web site
Outline
IntroductionWeb MiningWeb Content MiningWeb Structure MiningWeb Usage MiningConclusion & Exam Questions
Web Usage MiningTries to predict user behavior from interaction with the WebWide range of data (logs)
Web client data Proxy server data Web server dataTwo common approaches
1. Map usage data into relational tables before using adapted data mining techniques
2. Use log data directly by utilizing special pre-processing techniques
Web Usage Mining
Typical problems: Distinguishing among unique users, server sessions, episodes, etc in the presence of caching and proxy serversOften Usage Mining uses some background or domain knowledge E.g. site topology, Web content, etc
Web Usage Mining
Two main categories:1. Learning a user profile (personalized)
Web users would be interested in techniques that learn their needs and preferences automatically
2. Learning user navigation patterns (impersonalized)
Information providers would be interested in techniques that improve the effectiveness of their Web site or biasing the users towards the goals of the site
Outline
IntroductionWeb MiningWeb Content MiningWeb Structure MiningWeb Usage MiningConclusion & Exam Questions
Conclusions
Tried to resolve confusion with regards to the term Web Mining Differentiated from IR and IE
Suggest three Web mining categories: Content, Structure, and Usage Mining
Briefly described approaches for the three categoriesExplored connection with agent paradigm
Exam Question #1
Question: Outline the main characteristics of Web information.
Answer: Web information is huge, diverse, and dynamic.
Exam Question #2Question: How data mining techniques can be used in Web information analysis? Give at least two examples. Classification: classification on server logs using decision
tree, Naïve-Bayes classifier to discover the profiles of users belonging to a particular class
Clustering: Clustering can be used to group users exhibiting similar browsing patterns.
Association Analysis: association analysis can be used to relate pages that are most often referenced together in a single server session.
Exam Question #3
Question: What are the three main areas of interest for Web mining?
Answer: (1) Web Content (2) Web Structure
(3) Web Usage
Thank you!