Social Tagging and Search Marti Hearst UC Berkeley.

Slides:



Advertisements
Similar presentations
Haystack: Per-User Information Environment 1999 Conference on Information and Knowledge Management Eytan Adar et al Presented by Xiao Hu CS491CXZ.
Advertisements

ONLINE RESOURCES. QUESTION Do you ever go onto the Internet and plan to only spend a small amount of time looking for something and spend much longer.
ONLINE RESOURCES. QUESTION Do you ever go into the Internet and plan to only spend a small amount of time looking for something and spend much longer.
Bringing Order to the Web: Automatically Categorizing Search Results Hao Chen SIMS, UC Berkeley Susan Dumais Adaptive Systems & Interactions Microsoft.
Personalization and Search Jaime Teevan Microsoft Research.
Sean Blong Presents: 1. What are they…?  “[…] specific type of information filtering (IF) technique that attempts to present information items (movies,
Information retrieval mon jan data…. framework for today’s lecture…
1 Learning User Interaction Models for Predicting Web Search Result Preferences Eugene Agichtein Eric Brill Susan Dumais Robert Ragno Microsoft Research.
Tagging Systems Austin Wester. Tags A keywords linked to a resource (image, video, web page, blog, etc) by users without using a controlled vocabulary.
Semi-Automated Creation of Facet Hierarchies Marti Hearst School of Information, UC Berkeley Joint work with Dr. Emilia Stoica.
Basic IR: Queries Query is statement of user’s information need. Index is designed to map queries to likely to be relevant documents. Query type, content,
Castanet: Using WordNet to Build Facet Hierarchies Emilia Stoica and Marti Hearst School of Information, Berkeley.
Measuring Information Architecture CHI 01 Panel Position Statement Marti Hearst UC Berkeley.
Faceted Metadata for Site Navigation and Search Marti Hearst 12/17/2009.
Measuring Information Architecture Marti Hearst UC Berkeley.
Measuring Information Architecture Marti Hearst UC Berkeley.
Semi-Automated Creation of Facet Hierarchies Marti Hearst School of Information, UC Berkeley Joint work with Dr. Emilia Stoica.
A metadata-based approach Marti Hearst Associate Professor BT Visit August 18, 2005.
Yahoo Visit Day Joint Reseach Opportunities Marti Hearst UC Berkeley School of Information.
Best Practices for Search for the Federal Government Marti Hearst Web Manager University November 10, 2009.
1 CS 430 / INFO 430 Information Retrieval Lecture 24 Usability 2.
SEO: Past, Present, Future Name Company Twitter. SEO Tips from Website Grader Lessons from 2,602,042 websites.
1 i247: Information Visualization and Presentation Marti Hearst April 7, 2008.
Faceted Metadata for Information Architecture and Search Marti Hearst, SIMS at UC Berkeley Preston Smalley & Corey Chandler, eBay User Experience & Design.
Future Trends in Search User Interfaces Dr. Marti Hearst UC Berkeley i-Know Conference Keynote Sept 1, 2010.
Thoughts on Tagging & Search Marti Hearst UC Berkeley.
Usability of Grouping of Retrieval Results Marti Hearst School of Information, UC Berkeley September 1, 2006.
UIs for Faceted Navigation Recent Advances and Remaining Open Problems HCIR’08 Marti Hearst, UC Berkeley (including some slides from Corey Chandler of.
1 User Interfaces for Information Access Marti Hearst IS202, Fall 2006.
Measuring Information Architecture Marti Hearst UC Berkeley.
1 Next-Level Discovery Panel Marti Hearst UC Berkeley.
Emerging Trends in Search User Interfaces Prof. Marti Hearst UC Berkeley PSU Graduate Research Symposium March 25, 2011 Book full text freely available.
Ideas for USA.gov Marti Hearst USA.gov & Web Best Practices Team Meeting July 29, 2009.
Library 10 – Information Competency Search Engines.
Information retrieval thur jan data…. framework for today’s lecture…
What difference a good tool? using Endeca for a faceted catalog Emily Lynema NCSU Libraries ACRL Delaware Valley Chapter Fall Program November 3, 2006.
+ Social Bookmarking and Collaborative Filtering Christopher G. Wagner.
CS598CXZ Course Summary ChengXiang Zhai Department of Computer Science University of Illinois, Urbana-Champaign.
Web 2.0: Concepts and Applications 4 Organizing Information.
Personalization of the Digital Library Experience: Progress and Prospects Nicholas J. Belkin Rutgers University, USA
Web 2.0: Concepts and Applications 6 Linking Data.
Bringing Order to the Web: Automatically Categorizing Search Results Hao Chen, CS Division, UC Berkeley Susan Dumais, Microsoft Research ACM:CHI April.
Information retrieval wed sept data…. -start at 6.45.
Web Data Management Dr. Daniel Deutch. Web Data The web has revolutionized our world Data is everywhere Constitutes a great potential But also a lot of.
1 Mining User Behavior Mining User Behavior Eugene Agichtein Mathematics & Computer Science Emory University.
OCLC Research OCLC Online Computer Library Center Research & New Technologies Interest Group 24 October 2005 DeweyBrowser & Curiouser Diane Vizine-Goetz.
Topical Crawlers for Building Digital Library Collections Presenter: Qiaozhu Mei.
SUMMON ® 2.0 DISCOVERY REINVENTED. What is Summon 2.0? A new, streamlined, modern interface New and enhanced features providing layers of contextual guidance.
Search Result Interface Hongning Wang Abstraction of search engine architecture User Ranker Indexer Doc Analyzer Index results Crawler Doc Representation.
Faceted Search Zhao Jing Outline  What is faceted search?  Why use faceted search?  Topics of interests  Faceted Search in Dataspace.
How can Search Interfaces Enhance the Value of Semantic Annotations (and Vice Versa?) Keynote Talk ESAIR’13: Sixth International Workshop on Exploiting.
Collaborative Information Retrieval - Collaborative Filtering systems - Recommender systems - Information Filtering Why do we need CIR? - IR system augmentation.
Natural Search User Interfaces Prof. Marti Hearst UC Berkeley March/April, 2012 Book full text freely available at:
Searching the web Enormous amount of information –In 1994, 100 thousand pages indexed –In 1997, 100 million pages indexed –In June, 2000, 500 million pages.
SEO Who knew 3 letters could mean so much?. What is SEO? Search Engine Optimization (SEO) is the practice of improving and promoting a web site in order.
Understanding User Goals in Web Search University of Seoul Computer Science Database Lab. Min Mi-young.
WEB 2.0 PATTERNS Carolina Marin. Content  Introduction  The Participation-Collaboration Pattern  The Collaborative Tagging Pattern.
Working Memory and Learning Underlying Website Structure
Search Result Interface Hongning Wang Abstraction of search engine architecture User Ranker Indexer Doc Analyzer Index results Crawler Doc Representation.
Juhaida Abdul Aziz Parilah M Shah Rosseni Din Rashidah Rahamat.
KMS & Collaborative Filtering Why CF in KMS? CF is the first type of application to leverage tacit knowledge People-centric view of data Preferences matter.
Bringing Order to the Web : Automatically Categorizing Search Results Advisor : Dr. Hsu Graduate : Keng-Wei Chang Author : Hao Chen Susan Dumais.
Summon® 2.0 Discovery Reinvented
NLP Support for Faceted Navigation in Scholarly Collections
Web Engineering.
USERS’ PERCEPTIONS OF THE E-MENU PROTOTYPE ON E-MENU FEATURES
Document Clustering Matt Hughes.
Incorporating Metadata into Search User Interfaces
Moodle Training — Advanced Topics —
A Glimpse of Recommender Systems on the Web
Presentation transcript:

Social Tagging and Search Marti Hearst UC Berkeley

2 Marti Hearst, iConfernece ‘06 Search Topical Metadata Structured, Flexible Navigation

3 Marti Hearst, iConfernece ‘06 The Idea of Facets  Create INDEPENDENT categories (facets)  Each facet has labels (sometimes arranged in a hierarchy)  Assign labels from the facets to every item  Example: recipe collection Course Main Course Cooking Method Stir-fry Cuisine Thai Ingredient Bell Pepper Curry Chicken

4 Marti Hearst, iConfernece ‘06 Using Facets  Allow multiple ways to get to each item Preparation Method Fry Saute Boil Bake Broil Freeze Desserts Cakes Cookies Dairy Ice Cream Sherbet Flan Fruits Cherries Berries Blueberries Strawberries Bananas Pineapple Fruit > Pineapple Dessert > Cake Preparation > Bake Dessert > Dairy > Sherbet Fruit > Berries > Strawberries Preparation > Freeze

5 Marti Hearst, iConfernece ‘06 Opening View Select literature from PRIZE facet

6 Marti Hearst, iConfernece ‘06 Group results by YEAR facet

7 Marti Hearst, iConfernece ‘06 Select 1920’s from YEAR facet

8 Marti Hearst, iConfernece ‘06 Current query is PRIZE > literature AND YEAR: 1920’s. Now remove PRIZE > literature

9 Marti Hearst, iConfernece ‘06 Now Group By YEAR > 1920’s

10 Marti Hearst, iConfernece ‘06 Advantages of the Approach  Systematically integrates search results:  reflect the structure of the info architecture  retain the context of previous interactions  Gives users control and flexibility  Over order of metadata use  Over when to navigate vs. when to search  Allows integration with advanced methods  Collaborative filtering, predicting users’ preferences

11 Marti Hearst, iConfernece ‘06 Faceted Digital Libraries  NCSU has a start at it

Problem with Metadata-Oriented Approaches Getting the metadata!

13 Marti Hearst, iConfernece ‘06 Search Topical Metadata Social question answering Recorded Human Interaction Click-through ranking Inferred recommendations

14 Marti Hearst, iConfernece ‘06 Human Real-time Question Answering  More popular in Korea than algorithmic search  Maybe fewer good web pages?  Maybe more social society?  Several examples in US:  Yahoo answers recently released and successful  wondir.com  answerbag.com

15 Marti Hearst, iConfernece ‘06 Yahoo Answers (also answerbag.com, wondir.com, etc)

16 Marti Hearst, iConfernece ‘06 Yahoo Answers appearing in search results

17 Marti Hearst, iConfernece ‘06 answerbag.com

18 Marti Hearst, iConfernece ‘06 Using User Behavior as Implicit Preferences  Search click-through experimentally shown to boost search rankings for top results  Joachims et al. ‘05, Agichtein et al. ‘06  Works ok even if non-relevant documents examined  Best in combination with sophisticated search algorithms  Doesn’t work well for ambiguous queries  Aggregates of movie and book selections comprise implicit recommendations

19 Marti Hearst, iConfernece ‘06 Search Topical Metadata Recorded Human Interaction Social Tagging (photos, bookmarks) Game-based tagging

20 Marti Hearst, iConfernece ‘06 Social Tagging  Metadata assignment without all the bother  Spontaneous, easy, and tends towards single terms

21 Marti Hearst, iConfernece ‘06 Issues with Photo and Web link Tagging  There is a strong personal component  Marking for my own reminders  Marking for my circle of friends  There is also a strong social component  Try to promote certain tags to make them more popular, or post to popular tags to see your influence rise

22 Marti Hearst, iConfernece ‘06 Tagging Games  Assigning metadata is fun! (ESP game, von Ahn)  No need for reputation system, etc.  Pay people to do it  MyCroft (iSchool student project)  Drawback: least common denominator labels  Experts already label their own data or that about which they have expertise  E.g., protein function  Wikipedia

23 Marti Hearst, iConfernece ‘06 Search Topical Metadata Social question answering Recorded Human Interaction Social Tagging (photos, bookmarks) Click-through ranking Inferred recommendations Game-based tagging ????

24 Marti Hearst, iConfernece ‘06 Expert-Oriented Tagging in Search  Already happening at Google co-op  Shows up in certain types of search results

25 Marti Hearst, iConfernece ‘06 Expert-Oriented Tagging  Already happening at Google co-op  Shows up in certain types of search results

26 Marti Hearst, iConfernece ‘06 Promoting Expertise-Oriented Tagging  Research area: User Interfaces  To make rapid-feedback suggestions of pre- established tags  Like type-ahead queries  To incentivize labeling and make it fun  To allow the personal aspects to shine through

27 Marti Hearst, iConfernece ‘06 Promoting Expertise-Oriented Tagging  Research area: NLP Algorithms  (We have an algorithm to build facets from text)  To convert tags into facet hierarchies  To capture implicit labeling information

28 Marti Hearst, iConfernece ‘06 Promoting Expertise-Oriented Tagging  Research area: Digital infrastructure  Extending tagging games  Build an architecture that channels specialized subproblems to appropriate experts  We now know there is a green plant in an office; direct this to the botany > houseplants experts

29 Marti Hearst, iConfernece ‘06 Promoting Expertise-Oriented Tagging  Research area: economics and sociology  What are the right incentive structures?

30 Marti Hearst, iConfernece ‘06 Using Implicit Preferences  Extend implicit recommendation technology to online catalog use

31 Marti Hearst, iConfernece ‘06 Summary  There is great potential in tapping the social information use channel  To improve metadata  To improve integration with search  The necessary research is interdisciplinary!