Knowledge Discovery in a multi-mode network: a visualization approach to measure the interdependence between actors and social venues Dragos Calitoiu Zachary.

Slides:



Advertisements
Similar presentations
Peer-to-Peer and Social Networks Power law graphs Small world graphs.
Advertisements

Complex Networks for Representation and Characterization of Images For CS790g Project Bingdong Li 9/23/2009.
4.7 Graphing Lines Using Slope Intercept Form
Schema Summarization cong Yu Department of EECS University of Michigan H. V. Jagadish Department of EECS University of Michigan
Feb 20, Definition of subgroups Definition of sub-groups: “Cohesive subgroups are subsets of actors among whom there are relatively strong, direct,
Small-World Graphs for High Performance Networking Reem Alshahrani Kent State University.
Funding Networks Abdullah Sevincer University of Nevada, Reno Department of Computer Science & Engineering.
Using Structure Indices for Efficient Approximation of Network Properties Matthew J. Rattigan, Marc Maier, and David Jensen University of Massachusetts.
Search in a Small World JIN Xiaolong Based on [1].
CSE 222 Systems Programming Graph Theory Basics Dr. Jim Holten.
3.7 Graphs of Rational Functions
Design and Specifications of a Visualization Tool for Dynamic Network Analysis (VITA-DNA) Marcus Lem, MD, MHSc, FRCPC Health Canada Ben Houston Exocortex.
Robustness of complex networks English workshop on LEAP May 8, 2012 Kyushu Univ. Yuki YOSHIDA Dept. AEEA. Kyushu Univ.
Introduction to Data Mining Group Members: Karim C. El-Khazen Pascal Suria Lin Gui Philsou Lee Xiaoting Niu.
Efficient Identification of Overlapping Communities Jeffrey Baumes Mark Goldberg Malik Magdon-Ismail Rensselaer Polytechnic Institute, Troy, NY.
Small World Social Networks With slides from Jon Kleinberg, David Liben-Nowell, and Daniel Bilar.
Principles of Social Network Analysis. Definition of Social Networks “A social network is a set of actors that may have relationships with one another”
VITA for links and nodes analysis network analysis for counterror & intelligence Zachary Jacobson, Health Canada with Ben Houston and Olivier Dagenais,
Clustering Spatial Data Using Random Walks Author : David Harel Yehuda Koren Graduate : Chien-Ming Hsiao.
Science: Graph theory and networks Dr Andy Evans.
Random-Graph Theory The Erdos-Renyi model. G={P,E}, PNP 1,P 2,...,P N E In mathematical terms a network is represented by a graph. A graph is a pair of.
An Introduction to Social Network Analysis Yi Li
A Graph-based Friend Recommendation System Using Genetic Algorithm
Mining Social Networks for Personalized Prioritization Shinjae Yoo, Yiming Yang, Frank Lin, II-Chul Moon [KDD ’09] 1 Advisor: Dr. Koh Jia-Ling Reporter:
Lecture 13: Network centrality Slides are modified from Lada Adamic.
3. SMALL WORLDS The Watts-Strogatz model. Watts-Strogatz, Nature 1998 Small world: the average shortest path length in a real network is small Six degrees.
Social Networks. 2 A social network is a social structure made up of individuals or organizations (called "nodes“), which are tied (connected) by one.
Discrete Structures Graphs 1 (Ch. 10) Dr. Muhammad Humayoun Assistant Professor COMSATS Institute of Computer Science, Lahore.
ROCK: A Robust Clustering Algorithm for Categorical Attributes Authors: Sudipto Guha, Rajeev Rastogi, Kyuseok Shim Data Engineering, Proceedings.,
To add fractions, you need a common denominator. Remember!
Slides are modified from Lada Adamic
The author wants you to FOIL! STOP! TOO MUCH WORK! Alternative, less work! Combine both complex zeros with + sign. Isolate the i term. Square both sides.
Chapter P Prerequisites: Fundamental Concepts of Algebra 1 Copyright © 2014, 2010, 2007 Pearson Education, Inc. 1 P.7 Equations.
March 3, 2009 Network Analysis Valerie Cardenas Nicolson Assistant Adjunct Professor Department of Radiology and Biomedical Imaging.
Small World Social Networks With slides from Jon Kleinberg, David Liben-Nowell, and Daniel Bilar.
1 Finding Spread Blockers in Dynamic Networks (SNAKDD08)Habiba, Yintao Yu, Tanya Y., Berger-Wolf, Jared Saia Speaker: Hsu, Yu-wen Advisor: Dr. Koh, Jia-Ling.
Relation Strength-Aware Clustering of Heterogeneous Information Networks with Incomplete Attributes ∗ Source: VLDB.
Networks are connections and interactions. Networks are present in every aspect of life. Examples include economics/social/political sciences. Networks.
Informatics tools in network science
Lines that a function approaches but does NOT actually touch.
Breaking Attendance Barriers Spark 2015 Trinidad & Tobago District.
Paper Presentation Social influence based clustering of heterogeneous information networks Qiwei Bao & Siqi Huang.
EXAMPLE FORMULA DEFINITION 1.
Graph clustering to detect network modules
Bell Ringer Solve even #’s.
Presenter: Waqas Nawaz
MTH1150 Tangents and Their Slopes
Social Networks Analysis
3-3: Cramer’s Rule.
Graphing Linear Equations
Fourier Series.
Comparison of Social Networks by Likhitha Ravi
2.1 Equations of Lines Write the point-slope and slope-intercept forms
Graphing Linear Equations
The Watts-Strogatz model
Computer Vision Lecture 9: Edge Detection II
Effective Social Network Quarantine with Minimal Isolation Costs
Small World Networks Scotty Smith February 7, 2007.
Department of Computer Science University of York
Precalculus Essentials
Peer-to-Peer and Social Networks
Unit 6: Ratios: SPI : Solve problems involving ratios, rates, and percents Remember to have paper and pencil ready at the beginning of each.
Discovery of Blog Communities based on Mutual Awareness
Korea University of Technology and Education
Local Clustering Coefficient
Malik Magdon-Ismail, Konstantin Mertsalov, Mark Goldberg
Solving Equations Review
Alternative, less work! Combine both complex zeros with + sign.
Demo data transformation
Linear Functions and Slope-Intercept Form Lesson 2-3
Presentation transcript:

Knowledge Discovery in a multi-mode network: a visualization approach to measure the interdependence between actors and social venues Dragos Calitoiu Zachary Jacobson Margaret Varga Ben Houston Health Canada, Exocortex, QinetiQ, Carleton Univ. Nov. 2008

Abstract: We propose: - separating the set of actors and their personal connections in a social network from the sets of superordinate and subordinate connections among them [e.g., places, organizations, events]. - building a structure that contains the network of actors, the network of other properties and the links between them. With this separation, we propose measures to describe how the network of actors affects the other networks, and also to describe how the network of places influences the structure of the network of actors.

fx13 f030 f008 f103 f104 m107 f006 m106 f002 m112 mx04 m211 f021 m212 m018 f023 fx21 m301 f024 m200 m546 mx06 f541 m014 f013 m523 mx01 m102 mx05 fx12 f536 m101 f007 fx21 mx11 fx03 mx12 mx10 m306 f029 m304 m537

f010 m026 f015 m017 f034 fx36 f033 f009 m201 fx07 m202 m012 f035 m013 f201 m551 f900 f514 m526 m206 f019 f202 f017 m207 f012 f020 m209 f533 f014 f022 m203 m204 f011 m016 f038 m019 m210 m208 m025 f002f030 m106 f006 m107 f104 f103 fx13 f008 m112 m007 mx06f541 mx04 m102 mx05 f013 fx12 f004 m014 mx01 f536 m523 m101 f546 m212 fx06 fx21 m301 f024 m200 mx14 m302 f025 f021 m018 f023 m211 m002 m010 f106 f007 m111m110 m023 f205 m214 f016 f003 m104 Sexual network member Member attending bar Bar

One-mode visualization [ We are very grateful to Dr. Ann Jolly who provided us the data.]

3D visualization (iteration 0) There are two persons (#’s 96 and 98), each linked with the same two places. We show a link between these two places in plane P2.

3D visualization (iteration 1) If two individuals, who are directly connected with two different palaces, are also connected in P1 (neighborhood of order 1, namely directly connected), we show a link in P2 between these two places.

3D visualization (iteration 2) If two individuals who are direct connected with two different places are also connected in P1 (neighborhood of order 2), we show a link in P2 between these two places.

New local measures k-Local Place Connectedness (k-LPC) k-Local Link Connectedness (k-LLC) k-Local Ratio of Connectedness (k-LRC k-Local Place Efficiency of order i (k-LPE of order i) k-Local Links Efficiency of order i: (k-LLE of order i) (Definitions in the paper)

New global measures Global Place Connectedness (GPC) Global Link Connectedness (GLC) Global Ratio of Connectedness (GRC) Global Place Efficiency of order i (GPE of order i) Global Links Efficiency of order i (GLE of order i) (Definitions published for 2 levels)

Generalization to multi-modes Rigorous generalized measures under development and study

Clustering coefficient – a new approach for N-mode network Classical definitions: The local clustering coefficient for an individual node i with deg i neighbors and  i edges between its neighbors is: an alternative and more widely used definition of the clustering coefficient This formula is basically not defined if the number of neighbors deg i becomes zero or one as the denominator becomes zero. These cases are usually treated as C i = 0 although some authors also set these values to one.

Clustering coefficient – a new approach for N-mode network Solving the definition: Marcus Kaiser (Newcastle University) looked at the effect of removing nodes with less than two neighbors corresponding to leafs and isolated nodes before averaging for the global clustering coefficient. The relation between the new coefficient C’ and the traditional measure C 1 can be derived from the fraction of nodes that have one or zero neighbors,  by A new computational problem for a N-mode network

Questions and Comments: - What next? - Now what?