University of Buffalo The State University of New York Spatiotemporal Data Mining on Networks Taehyong Kim Computer Science and Engineering State University.

Slides:



Advertisements
Similar presentations
Complex Networks Advanced Computer Networks: Part1.
Advertisements

Distributed Advice-Seeking on an Evolving Social Network Dept Computer Science and Software Engineering The University of Melbourne - Australia Golriz.
Epidemics How can we protect ourselves against bird flu?
Modeling Malware Spreading Dynamics Michele Garetto (Politecnico di Torino – Italy) Weibo Gong (University of Massachusetts – Amherst – MA) Don Towsley.
Wildlife Disease Avian Influenza John F. Corbett, III Bio. 335-Wildlife and Fisheries Biology Keystone College Keystone College Feb. 18, 2010.
Movement-Assisted Sensor Deployment Author : Guiling Wang, Guohong Cao, Tom La Porta Presenter : Young-Hwan Kim.
School of Information University of Michigan Network resilience Lecture 20.
Mmmmm Mohamed M. B. Alnoor CHP400 COMMUNITY HEALTH PROGRAM-II Avian Influenza H5N1 Epidemiology & Control mmmmm.
Edith C. H. Ngai1, Jiangchuan Liu2, and Michael R. Lyu1
Presentation Topic : Modeling Human Vaccinating Behaviors On a Disease Diffusion Network PhD Student : Shang XIA Supervisor : Prof. Jiming LIU Department.
SFU, CMPT 741, Fall 2009, Martin Ester 418 Outlook Outline Trends in KDD research Graph mining and social network analysis Recommender systems Information.
Scale Free Networks Robin Coope April Abert-László Barabási, Linked (Perseus, Cambridge, 2002). Réka Albert and AL Barabási,Statistical Mechanics.
Networks FIAS Summer School 6th August 2008 Complex Networks 1.
1 Epidemic Spreading in Real Networks: an Eigenvalue Viewpoint Yang Wang Deepayan Chakrabarti Chenxi Wang Christos Faloutsos.
1 PUBLIC - PRIVATE PARTNERSHIP FOR AVIAN INFLUENZA CONTROL AND PANDEMIC PREPAREDNESS Presented by Bayu Krisnamurthi Indonesia National Committee for Avian.
Internet Quarantine: Requirements for Containing Self- Propagating Code David Moore, Colleen Shannon, Geoffrey M. Voelker, Stefan Savage.
1 Avian Influenza Rapid Response Team Training. 2 What is a Rapid Response Team? A team of professionals that investigates suspected cases of avian influenza.
Avian Influenza – What does it all mean? Important Background Information Island Paravets and Residents.
Modelling the control of epidemics by behavioural changes in response to awareness of disease Savi Maharaj (joint work with Adam Kleczkowski) University.
Triangulation of network metaphors The Royal Netherlands Academy of Arts and Sciences Iina Hellsten & Andrea Scharnhorst Networked Research and Digital.
Systems Biology, April 25 th 2007Thomas Skøt Jensen Technical University of Denmark Networks and Network Topology Thomas Skøt Jensen Center for Biological.
Internet Quarantine: Requirements for Containing Self-Propagating Code David Moore et. al. University of California, San Diego.
Pandemic Influenza Preparedness Kentucky Department for Public Health Department for Public Health.
Systematic Analysis of Interactome: A New Trend in Bioinformatics KOCSEA Technical Symposium 2010 Young-Rae Cho, Ph.D. Assistant Professor Department of.
How serious is the threat of an Avian flu Human Pandemic Avian (Bird) December 2005.
Models of Influence in Online Social Networks
Developing a vaccine and how a pandemic could occur.
Large-scale organization of metabolic networks Jeong et al. CS 466 Saurabh Sinha.
Epidemic spreading in complex networks: from populations to the Internet Maziar Nekovee, BT Research Y. Moreno, A. Paceco (U. Zaragoza) A. Vespignani (LPT-
The Global Epidemic Simulator Wes Hinsley 1, Pavlo Minayev 1 Stephen Emmott 2, Neil Ferguson 1 1 MRC Centre for Outbreak Analysis and Modelling, Imperial.
Course Overview & Introduction to Social Network Analysis How to analyse social networks?
Online Social Networks and Media Epidemics and Influence.
Demetris Kennes. Contents Aims Method(The Model) Genetic Component Cellular Component Evolution Test and results Conclusion Questions?
Soon-Hyung Yook, Sungmin Lee, Yup Kim Kyung Hee University NSPCS 08 Unified centrality measure of complex networks.
Efficient Gathering of Correlated Data in Sensor Networks
SIR Epidemic Models CS 390/590 Fall 2009
Network Analysis Diffusion Networks. Social Network Philosophy Social structure is visible in an anthill Movements & contacts one sees are not random.
V5 Epidemics on networks
Epidemiology of Influenza. The Flu Basics The flu is contagious and can range from mild to deadly Each year between 5% and 20% of the US population contracts.
Code Red Worm Propagation Modeling and Analysis Cliff Changchun Zou, Weibo Gong, Don Towsley Univ. Massachusetts, Amherst.
1 Operational and Logistical Aspects of Biodefense Moshe Kress CEMA, Israel.
Clustering of protein networks: Graph theory and terminology Scale-free architecture Modularity Robustness Reading: Barabasi and Oltvai 2004, Milo et al.
A Graph-based Friend Recommendation System Using Genetic Algorithm
Hubba: Hub Objects Analyzer—A Framework of Interactome Hubs Identification for Network Biology 吳 信 宏, Hsin-Hung Wu Laboratory.
1 Worm Propagation Modeling and Analysis under Dynamic Quarantine Defense Cliff C. Zou, Weibo Gong, Don Towsley Univ. Massachusetts, Amherst.
Showcase /06/2005 Towards Computational Epidemiology Using Stochastic Cellular Automata in Modeling Spread of Diseases Sangeeta Venkatachalam, Armin.
Opportunities and Challenges in “Coupled Natural-Human Systems and Emerging Infectious Diseases in Vietnam” Dr. Trinh Dinh Thau, Vice Dean Faculty of Veterinary.
Pandemic Influenza: A Primer for Organizational Preparation Pandemic Influenza: A Primer for Organizational Preparation Kristine Perkins, MPH Director,
The Vermont Department of Health Overview of Pandemic Influenza Regional Pandemic Planning Summits 2006 Guidance Support Prevention Protection.
Page 1 Inferring Relevant Social Networks from Interpersonal Communication Munmun De Choudhury, Winter Mason, Jake Hofman and Duncan Watts WWW ’10 Summarized.
Robustness of complex networks with the local protection strategy against cascading failures Jianwei Wang Adviser: Frank,Yeong-Sung Lin Present by Wayne.
Network theory 101 Temporal effects What we are interested in What kind of relevant temporal /topological structures are there? Why? How does.
Genome Biology and Biotechnology The next frontier: Systems biology Prof. M. Zabeau Department of Plant Systems Biology Flanders Interuniversity Institute.
1 Epidemic Spreading Parameters: External Model based on population density and travel statistics.
A Spatial-Temporal Model for Identifying Dynamic Patterns of Epidemic Diffusion Tzai-Hung Wen Associate Professor Department of Geography,
Janine Bolliger Swiss Federal Research Institute WSL/FNP,
Day CREATING A WORLD THAT IS SAFE AND SUSTAINABLE FOR WILDLIFE AND SOCIETY Avian Influenza in Wild Birds Matching goals and methods.
Class 21: Spreading Phenomena PartI
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.
An Improved Acquaintance Immunization Strategy for Complex Network.
1 Lesson 12 Networks / Systems Biology. 2 Systems biology  Not only understanding components! 1.System structures: the network of gene interactions and.
Algorithms and Computational Biology Lab, Department of Computer Science and & Information Engineering, National Taiwan University, Taiwan Network Biology.
Trends and dynamics of HPAI - epidemiological and animal health risks Technical Meeting on HPAI and Human H5N1 Infection Rome, Italy, June 27-29, 2007.
Sangeeta Venkatachalam, Armin R. Mikler
Network Science in NDSSL at Virginia Tech
Effective Social Network Quarantine with Minimal Isolation Costs
Department of Computer Science University of York
Topology and Dynamics of Complex Networks
Regulation Analysis using Restricted Boltzmann Machines
Susceptible, Infected, Recovered: the SIR Model of an Epidemic
Presentation transcript:

University of Buffalo The State University of New York Spatiotemporal Data Mining on Networks Taehyong Kim Computer Science and Engineering State University of New York at Buffalo

University of Buffalo The State University of New York Table of Contents  Studies  Spreading and Defense model in Networks  Fixed-random network  Spreading Model  Defense Model  Avian Influenza Outbreaks  Modeling  Mining parameters  Introduction  Overview  Networks Data Mining  Spatiotemporal Data Mining  Applications  Quality of Bone (osteoporosis) as a Network Dynamics  Amazon Deforestation

University of Buffalo The State University of New York Overview  Most of real world relationships and communications could be represented on networks (graphs).  Understanding the behavior of such systems starts with understanding the topology of the corresponding network. Yeast PPI networkAT&T Web NetworkCollaboration network

University of Buffalo The State University of New York Overview  Recent studies on various networks  Social network  Author network, School relationship Network  Technical network  Cell network, Internet, Electric power network  Biological network  Protein network, Metabolic network, Disease Network  Focuses on network attributes  Number of nodes and edges  Weight on nodes and edges

University of Buffalo The State University of New York Overview Hub node Bridge node edge node  nodes and edges

University of Buffalo The State University of New York Networks Data Mining  Networks Data mining has been done  Prediction of unknown protein functions in protein- protein interaction networks  Resilience test of networks against attacks  Prediction of people relationships in social networks  Drug targeting on cell networks  Etc.

University of Buffalo The State University of New York Spatiotemporal Data Mining  Networks are changed as time goes by  World wide web is evolving by itself  Interactions among proteins are changed in PPI networks  Size of cities and inter-state free ways are changed  Structure of bone is changed  Information of location and time is also important factors for further understanding on any given networks

University of Buffalo The State University of New York Spatiotemporal Data Mining  Spatiotemporal Data Mining: knowledge extraction from large spatiotemporal repositories in order to recognize behavioural trends and spatial patterns for prediction purposes  What is the relationship between the spread of epidemics and the number and location of houses and schools by time?  What is the connection between the size of Buffalo city and thruway traffics on I-90 by an year?

University of Buffalo The State University of New York Spatiotemporal Data Mining NormalOsteoporosis Drugs

University of Buffalo The State University of New York Amazon Deforestation 2003 Fonte: INPE PRODES Digital, Deforestation 2002/2003 Deforestation until 2002

University of Buffalo The State University of New York Amazon in 2015? fonte: Aguiar et al., 2004

University of Buffalo The State University of New York Modelling Complex Problems  Application of interdisciplinary knowledge to produce a model. If (... ? ) then... Desforestation?

University of Buffalo The State University of New York Table of Contents  Studies  Spreading and Defense model in Networks  Fixed-random network  Spreading Model  Defense Model  Avian Influenza Outbreaks  Modeling  Mining parameters  Introduction  Overview  Networks Data Mining  Spatiotemporal Data Mining  Applications  Quality of Bone (osteoporosis) as a Network Dynamics  Amazon Deforestation

University of Buffalo The State University of New York Spreading and Defense model in Networks  Fixed-radius random network  Cellular transmission tower  Interstate free ways  Epidemics on communities  Sensor networks  How we can defend if there are attacks or breaks from the center of the networks?

University of Buffalo The State University of New York Fixed Radius Random Network  400 random points on 1*1 square unit  Calculating distance between each point  If two points are in a certain radius, creating an edge between points

University of Buffalo The State University of New York Fixed Radius Random Network  Fixed-radius of random network (r = 0.01 ~ 0.14) Fixed-Radius 400 nodes, 2366 edges

University of Buffalo The State University of New York Simulation on network  Network dynamics are studied based on fixed-radius random network  Simple spreading model and defense model is implemented for simulation  Mining important parameters on this model of network dynamics  Mining optimal values of parameters on this model of network dynamics

University of Buffalo The State University of New York Spreading Model  Simulating disease spreading or message spreading  Starting from center point (0.5*0.5)  Affecting edges which are in a spreading radius (ROI) from center  Spreading radius grows or reduces based on how many edges are damaged

University of Buffalo The State University of New York Spreading Model  Region of radial distance of spreading model (ROI t=0 = 0.1)  Spreading starts from center (0.5, 0.5) ROI Center

University of Buffalo The State University of New York Spreading Model  Probability of affecting rate of edges (P a = 0.33)  11 edges are in ROI  In this case, 4 out of 11 edges are affected (Spreading will affect edges about 33% probability) ROI

University of Buffalo The State University of New York Defense Model  Simulating defense system of disease spreading or message spreading  Signaling to neighbor nodes in order to inform (disease) spreading  Activated when the affection of spreading (# of signals from neighbor nodes) is over threshold  Removing edges which are in a radius (  ) from activated neighbor nodes in order to stop spreading

University of Buffalo The State University of New York Defense Model  Circular region of programming Cell Death  0.2~3.6)  When signals from neighbor nodes are over the T d, edges in the circular region are removed by defense process Region of defense process

University of Buffalo The State University of New York Defense Model  Probability of Programming Cell Death (P p = 1)  If P p is 1, all edges in circular regions are dead

University of Buffalo The State University of New York Result (visualization) Time: 0Time: 10Time: 50 Total Damage Intermediate Contained Time 

University of Buffalo The State University of New York Result

University of Buffalo The State University of New York Result

University of Buffalo The State University of New York Summary  Containment strategy on epidemics and virus spreads  Mining important parameters  Mining optimal values of important parameters  Understanding dynamics on human tissues and bones  Development of diseases (osteoporosis)  Drug effects on cell networks

University of Buffalo The State University of New York Table of Contents  Studies  Spreading and Defense model in Networks  Fixed-random network  Spreading Model  Defense Model  Avian Influenza Outbreaks  Modeling  Mining parameters  Introduction  Overview  Networks Data Mining  Spatiotemporal Data Mining  Applications  Quality of Bone (osteoporosis) as a Network Dynamics  Amazon Deforestation

University of Buffalo The State University of New York Avian Influenza  AI outbreaks are frequently occurring around the world recently  H5N1 type has high infection and mortality rate  Chickens and ducks are main victims of AI  Mortality rate of H5N1 could reach % within 48 hours  Threat from AI has greatly increased for human beings  There are several reports showing human infection of AI  People could get infected by contacting excretion of contaminated birds

University of Buffalo The State University of New York AI outbreaks  Outbreaks in South Korea 2008

University of Buffalo The State University of New York AI outbreaks  Outbreaks in South Korea days 12 days 20 days 28 days 36 days 44 days

University of Buffalo The State University of New York Challenges  Strategies are needed for AI containment  Early identification of the first cluster of cases  Warning system from contaminated area to neighbor areas are needed  Effective quarantine plan should be existed  Containment model helps plan effective strategies  Prediction of damage with certain environment parameters  Mining important parameters to control outbreaks  Measurement of effective values of important parameters

University of Buffalo The State University of New York  A group of chickens and ducks are nodes  2231 nodes for a group of chickens and 808 nodes for a group of ducks  76 (1x1 square) units (1 unit = 37.5 Km)  Parameters  A node can interact with other nodes in range   A susceptible node become a infected node by infection probability   A Infected node become a activated node by incubation period  and   Nodes are culled in quarantine radius Modeling

University of Buffalo The State University of New York Modeling 487.5Km 300Km 37.5Km

University of Buffalo The State University of New York Visualization  Visualization of simulations based on AI outbreaks in South Korea days 14 days 24 days 34 days 44 days

University of Buffalo The State University of New York Important Parameters  Effect of Increased Quarantine Range  Quarantine radius: 0.0 ~ 0.32 unit  Effects of Increased Incubation Period  Incubation Period: 0 ~ 17 days  Effects of Increasing the Infection probability  Infection probability: 0.0 ~ 1.0

University of Buffalo The State University of New York Quarantine Radius  Effect of Increased Quarantine Radius  Quarantine radius: 0.0 ~ 0.32 unit  Infection probability: 0.1, 0.4, 0.7 and 1.0  Research on effective quarantine radius by Infection probability  Optimal quarantine radius Infection Probability Optimal Radius

University of Buffalo The State University of New York Quarantine Radius

University of Buffalo The State University of New York Incubation Period  Effects of Increased Incubation Period  Incubation Period: 0 ~ 17 days  Quarantine Range: 0.0, 0.04, 0.11 and 0.18 unit  For mid level control, almost 89% of poultry farms are healthy when incubation period is one day whereas only 11% of poultry farms are healthy when incubation period is 17 days.

University of Buffalo The State University of New York Infection probability  Effects of Increasing the Infection probability  Infection probability: 0.0 ~ 1.0  Quarantine Range: 0.0, 0.04, 0.11 and 0.18 unit  The large numbers of poultry farms eliminated by the aggressive culling procedure with max control

University of Buffalo The State University of New York Summary  Modeling AI dynamics based on statistic data  Modeling of AI outbreaks and spreads  Modeling of defense strategies  Mining important parameters and values in order to contain AI outbreaks in early stage  Quarantine radius, infection rate, incubation period  Damage predictions with important parameters  Mining defense strategies for future outbreaks

University of Buffalo The State University of New York Thank you!