The Complex Dynamics of Collaborative Tagging. Problem Tagging distributions tend to stabilize into powerlaw distributions. empirically determine as to.

Slides:



Advertisements
Similar presentations
Universal and Domain-specific Classifications from an Interdisciplinary Perspective Rick Szostak University of Alberta, Canada.
Advertisements

Topologies of Complex Networks Functions vs. Structures Lun Li Advisor: John C. Doyle Co-advisor: Steven H. Low Collaborators: David Alderson (NPS) Walter.
Chapter 5: Introduction to Information Retrieval
Power Laws By Cameron Megaw 3/11/2013. What is a Power Law?
Analysis and Modeling of Social Networks Foudalis Ilias.
Universal Search and Social Networking Exploiting the features of each to enhance the other and the tools that make it possible Peter Wallqvist Ravn Systems.
The Complex Dynamics of Collaborative Tagging Harry Halpin University of Edinburgh Valentin Robu CWI, Netherlands Hana Shepherd Princeton University WWW.
Emergence of Scaling in Random Networks Barabasi & Albert Science, 1999 Routing map of the internet
© 2007, Roman Schmidt Distributed Information Systems Laboratory Evergrow workshop, Jerusalem, IsraelFebruary 19, 2007 Efficient implementation of BP in.
PROBLEM BEING ATTEMPTED Privacy -Enhancing Personalized Web Search Based on:  User's Existing Private Data Browsing History s Recent Documents 
INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION Conceptualization of Place via Spatial Clustering and Co- occurrence Analysis.
Lecture 14: Collaborative Filtering Based on Breese, J., Heckerman, D., and Kadie, C. (1998). Empirical analysis of predictive algorithms for collaborative.
Time-dependent Similarity Measure of Queries Using Historical Click- through Data Qiankun Zhao*, Steven C. H. Hoi*, Tie-Yan Liu, et al. Presented by: Tie-Yan.
Creating Concept Hierarchies in a Customer Self-Help System Bob Wall CS /29/05.
Web as Graph – Empirical Studies The Structure and Dynamics of Networks.
1 Individual and Social Behavior in Tagging Systems Elizeu Santos-Neto David Condon, Nazareno Andrade Adriana Iamnitchi, Matei Ripeanu 20th ACM International.
Modern Information Retrieval Chapter 2 Modeling. Can keywords be used to represent a document or a query? keywords as query and matching as query processing.
Analysis of the Internet Topology Michalis Faloutsos, U.C. Riverside (PI) Christos Faloutsos, CMU (sub- contract, co-PI) DARPA NMS, no
Mink Spaans What are the problems that need to be solved What is hard.
Ranking by Odds Ratio A Probability Model Approach let be a Boolean random variable: document d is relevant to query q otherwise Consider document d as.
By Ciro Cattuto, Vittorio Loreto, and Luciano Pietronero Semiotic dynamics and collaborative tagging Present by Diyue Bu.
Algorithms for Data Mining and Querying with Graphs Investigators: Padhraic Smyth, Sharad Mehrotra University of California, Irvine Students: Joshua O’
Personalized Ontologies for Web Search and Caching Susan Gauch Information and Telecommunications Technology Center Electrical Engineering and Computer.
Semantic Web Technologies Lecture # 2 Faculty of Computer Science, IBA.
Tag-based Social Interest Discovery
Social Networking and On-Line Communities: Classification and Research Trends Maria Ioannidou, Eugenia Raptotasiou, Ioannis Anagnostopoulos.
Mark Levene, An Introduction to Search Engines and Web Navigation © Pearson Education Limited 2005 Slide 9.1 Chapter 9 : Social Networks What is a social.
In search for patterns of user interaction for digital libraries Jela Steinerová Comenius University, Bratislava, Slovakia
DYNAMICS OF COMPLEX SYSTEMS Self-similar phenomena and Networks Guido Caldarelli CNR-INFM Istituto dei Sistemi Complessi
1 11 Subcarrier Allocation and Bit Loading Algorithms for OFDMA-Based Wireless Networks Gautam Kulkarni, Sachin Adlakha, Mani Srivastava UCLA IEEE Transactions.
Percolation in self-similar networks Dmitri Krioukov CAIDA/UCSD M. Á. Serrano, M. Boguñá UNT, March 2011.
Community Evolution in Dynamic Multi-Mode Networks Lei Tang, Huan Liu Jianping Zhang Zohreh Nazeri Danesh Zandi & Afshin Rahmany Spring 12SRBIAU, Kurdistan.
Xiaoying Gao Computer Science Victoria University of Wellington Intelligent Agents COMP 423.
Crosscutting Concepts Next Generation Science Standards.
A Graph-based Friend Recommendation System Using Genetic Algorithm
1 Self-stabilizing Algorithms and Frequency Assignment Problems.
Friends and Locations Recommendation with the use of LBSN By EKUNDAYO OLUFEMI ADEOLA
Summarization of XML Documents K Sarath Kumar. Outline I.Motivation II.System for XML Summarization III.Ranking Model and Summary Generation IV.Example.
Machine Learning Tutorial Amit Gruber The Hebrew University of Jerusalem.
Normal Distribution Links The Normal Distribution Finding a Probability Standard Normal Distribution Inverse Normal Distribution.
Thesis Proposal: Prediction of popular social annotations Abon.
1 Reserving Ranges and Acceptable Deviations CANE Fall 2005 Meeting Kevin Weathers FCAS, MAAA The Hartford This document is designed for discussion purposes.
Harvesting Social Knowledge from Folksonomies Harris Wu, Mohammad Zubair, Kurt Maly, Harvesting social knowledge from folksonomies, Proceedings of the.
CS 8751 ML & KDDData Clustering1 Clustering Unsupervised learning Generating “classes” Distance/similarity measures Agglomerative methods Divisive methods.
Institute for Complex Systems Simulation An agent-based framework for analysing insolvency resolution mechanisms for banks Bob De Caux, Markus Brede and.
Design of a Robust Search Algorithm for P2P Networks
15 Sep 2015 EunJeong Cheon i501: introduction to informatics Semiotic Dynamics and Collaborative Tagging Ciro Cattuto, Vittorio Loreto, and Luciano Pietronero.
Enhanced hypertext categorization using hyperlinks Soumen Chakrabarti (IBM Almaden) Byron Dom (IBM Almaden) Piotr Indyk (Stanford)
1 Patterns of Cascading Behavior in Large Blog Graphs Jure Leskoves, Mary McGlohon, Christos Faloutsos, Natalie Glance, Matthew Hurst SDM 2007 Date:2008/8/21.
Correlation between People’s Behaviors in Cyber World and Their Geological Position Lixiong Chen Jan 24 th, 2009.
An Adaptive User Profile for Filtering News Based on a User Interest Hierarchy Sarabdeep Singh, Michael Shepherd, Jack Duffy and Carolyn Watters Web Information.
Response network emerging from simple perturbation Seung-Woo Son Complex System and Statistical Physics Lab., Dept. Physics, KAIST, Daejeon , Korea.
CS 440 Database Management Systems Web Data Management 1.
Section 6.1 Confidence Intervals for the Mean (Large Samples) © 2012 Pearson Education, Inc. All rights reserved. 1 of 83.
GRAPH AND LINK MINING 1. Graphs - Basics 2 Undirected Graphs Undirected Graph: The edges are undirected pairs – they can be traversed in any direction.
Dynamic Network Analysis Case study of PageRank-based Rewiring Narjès Bellamine-BenSaoud Galen Wilkerson 2 nd Second Annual French Complex Systems Summer.
13 Trends That Will Drive SEO in 2016 Presented By, Chennaiseocompany
Adversarial Information System Tanay Tandon Web Enhanced Information Management April 5th, 2011.
Item-to-Item Recommender Network Optimization
A Viewpoint-based Approach for Interaction Graph Analysis
Groups of vertices and Core-periphery structure
Requirement Prioritization
Personalized Social Image Recommendation
Systems of Inequalities
Ungraded quiz Unit 5.
Information Retrieval
CS 440 Database Management Systems
Department of Computer Science University of York
Degree Distributions.
Graph and Link Mining.
Presentation transcript:

The Complex Dynamics of Collaborative Tagging

Problem Tagging distributions tend to stabilize into powerlaw distributions. empirically determine as to why this pattern occurs and try to determine patterns prior to a stable distribution. The central question involved is when and how a coherent classification system might emerge from distributed tagging?

Solution Analyse tagging behaviour and their emperical results show that it does follow a powerlaw distribution. Dist(Ti, Tj) = N(Ti, Tj) / sqrt( N(Ti) * N(Tj) ) Based on these similarities they construct a tag-tag correlation graph or network. Size of Node = frequency of the tag. Distance = inverse of similarity between tags.

Criticism The information value of a tag suffers over time. Still human intervention needed. (Adversarial Nature) Tag Spam. There is no correspondence between keywords and concepts and it is not possible to do so with few keywords. Probably a combo of explicit (people) and implicit (software) ranking methods to get better tags and more useful search results.

Relation to what we learned in class Bottom-up approach to taxonomies/ontologies. It is a flat hierarchy. User Regulation – difficult. Because the user still uses keywords. The long tail phenomena.