Representing Higher-order Dependencies in Networks: Hands-on Tasks

Slides:



Advertisements
Similar presentations
Data Mining and the Web Susan Dumais Microsoft Research KDD97 Panel - Aug 17, 1997.
Advertisements

XProtect® Web Client 1 Product presentation.
University of Chicago Department of Energy The Parallel and Grid I/O Perspective MPI, MPI-IO, NetCDF, and HDF5 are in common use Multi TB datasets also.
Open an internet browser such as internet explorer.
Community Detection in a Large Real-World Social Network Karsten Steinhaeuser Nitesh V. Chawla DIAL Research Group University of Notre.
Social Network Analysis: Tasks and Tools Steven Loscalzo and Lei Yu Department of Computer Science Watson School of Engineering and Applied Science State.
Objectives  Understand the purpose of the superuser account  Outline the key features of the Linux desktops  Navigate through the menus  Getting help.
1 Mike Sinclair Kalamazoo Area Mathematics & Science Center University of Notre Dame Random walks A tutorial for the tutorial!
An Introduction to Designing and Executing Workflows with Taverna Aleksandra Pawlik University of Manchester materials by Dr Katy Wolstencroft and Dr Aleksandra.
Introduction to Information Retrieval CS 5604: Information Storage and Retrieval ProjCINETViz by Maksudul Alam, S M Arifuzzaman, and Md Hasanuzzaman Bhuiyan.
ing to Make Learning Relevant
Linux Tutorial Bret E. Shaw ’03 Computer Science Union College.
1 A Static Analysis Approach for Automatically Generating Test Cases for Web Applications Presented by: Beverly Leung Fahim Rahman.
Virginia Tech College of Liberal Arts and Human Sciences School of Education Office of Education, Research, and Outreach 404 Wallace.
6 October Information Retrieval and Knowledge Management Within the VisIT Web Services Architecture.
Tutorial 1: An Introduction to Visual Basic.NET1 Tutorial 1 An Introduction to Visual Basic.NET.
University Of Palestine. Department of Information Technology.
Aim: What is ImageJ and Why do we use it? Day 1. Brief Background Released in 1997 by Wayne Rasband Developed at the National Institutes of Health Runs.
Data Mining By Dave Maung.
Copyright © 2008, SAS Institute Inc. All rights reserved. Interactive Analysis and Data Visualization Using JMP −Dara Hammond, Federal Systems Engineer.
Jongwook Woo CIS 528 Introduction to Big Data Science (Syllabus) Jongwook Woo, PhD California State University, LA Computer and Information.
© 2010 Pearson Addison-Wesley. All rights reserved. Addison Wesley is an imprint of Designing the User Interface: Strategies for Effective Human-Computer.
Mantid Manipulation and Analysis Toolkit for ISIS data.
Chapter 3: Test Your Proficiency Directions: Select a section to work on. Work out each problem on a piece of paper. Click to check your answer. For detailed.
Overview: Ethics 450 Congrats on your new app! This application will provide FDCCI participants with an easy way to track their data center consolidation.
A RESEARCH SUPPORT SYSTEM FRAMEWORK FOR WEB DATA MINING Jin Xu, Yingping Huang, Gregory Madey Department of Computer Science and Engineering University.
Shakertown Webquest Click on the tool bar below to navigate through this assignment. Home TaskProcessEvaluation Conclusion.
Overview In this tutorial you will: learn what Janison is understand how to navigate through your course identify communication features in Janison. An.
Web mining is the use of data mining techniques to automatically discover and extract information from Web documents/services
SP5 - Neuroinformatics 3DSomaMS Tutorial Computational Intelligence Group Technical University of Madrid.
TCP-IP: Sliding Window protocol Sameer --- Brief description about the concept (1-2 lines)‏
Reference Management Module I: Introduction By Rehema Chande-Mallya(PhD)
Web Standards Web Design – Sec 2-3 Part or all of this lesson was adapted from the University of Washington’s “Web Design & Development I” Course materials.
Intro to Ethical Hacking
SLA San Diego Fall Seminar Round Table Discussion
Leader Name (Passport name) Major Nationality Graduate period or
Leader Name (Passport name) Major Nationality Graduate period or
HTTP and Abstraction on the Internet
Preparation of ULAB EVAL part C
Pagerank and Betweenness centrality on Big Taxi Trajectory Graph
Web Standards Web Design – Sec 2-3
Spatial Thinking in Non-design Areas
CS 854: Advanced Topics in Operating Systems
Searching the Internet
Flow Charts & Psuedo Code
Illustrate It! Use the crayons and large paper to illustrate the water cycle with detailed labels on each of the stages.
Web Standards Web Design – Sec 2-3
Upgrading My Computers
Web Design and Development
Intro to Ethical Hacking
Data Mining: Concepts and Techniques Course Outline
Big Blue Button A Canvas Workshop
Objectives Overview Explain why computer literacy is vital to success in today’s world Define the term, computer, and describe the relationship between.
Using Data to Explore Global Health Issues
Temporary Capacity Restrictions: TCR WG & TCR tool
Discovery Education Video Streaming and Board Builder
The EU DataGrid Project Tutorial
(Discussion – The Internet and the World Wide Web)
Install MySQL Community Server and MySQL Workbench
Share What You Have Learned EP Session
INTRODUCTION TO Data Analysis.
CHAPTER 14: Information Visualization
Getting Started with Splunk Power User eLearning
Science Student Survey Science Student Survey
CS31 Discussion 1H Fall18: week 1
Welcome to Outdoor Education (6/10)
Who is Using your webSite?
Welcome to the Erin’s Page Site Tour
Jiwon Kim Steve Seitz Maneesh Agrawala
Computer Vision Project
Presentation transcript:

Representing Higher-order Dependencies in Networks: Hands-on Tasks KDD 2018 Tutorial Beyond Graph Mining: Higher-Order Data Analytics for Temporal Network Data Representing Higher-order Dependencies in Networks: Hands-on Tasks Jian Xu Data Strategies Group Citadel LLC i@jianxu.net Nitesh Chawla Department of Computer Science and Engineering University of Notre Dame nchawla@nd.edu Good afternoon, my name is Jian, I’m from University of Notre Dame. Today I will discuss an approach to represent higher-order dependencies in networks. Aug 22nd, 2018

Overview of hands-on tasks Synthesizing trajectories with known variable orders of dependencies Use BuildHON+ (parameter-free) to extract variable orders of dependencies and build HON Use HONVis to visualize and interactively explore the higher-order network of NYC taxi data

Synthesizing trajectories with known variable orders of dependencies

Synthesizing trajectories Why synthetic? We know exactly when, where, and what types of higher-order dependencies exist. We inject 1st, 2nd, 3rd orders to trajectories. Details illustrated in the appendix. Initial setup: 10,000 users navigating through 100 web pages. Web pages are organized as a 10x10 grid and numbered from 00 to 99. Every page has two out-links to the neighboring pages, one pointing right and one pointing down, with wrapping. Every user clicks through 100 pages by moving right or moving down, resulting in 1,000,000 records.

Synthesizing trajectories Why synthetic? We know exactly when, where, and what types of higher-order dependencies exist. We inject 1st, 2nd, 3rd orders to trajectories. Details illustrated in the appendix. ./data/SyntheticTrajectoriesVariableOrders.csv 1 million web clickstream synthetic data User# Website1 Website 2 Website3…

Use BuildHON+ to extract variable orders of dependencies and build HON

BuildHON+ how to run

BuildHON+ extracted higher-order rules data/SyntheticTrajectoriesVariableOrders_rules.csv Example of 3rd order rule extracted data/SyntheticTrajectoriesVariableOrders_network.csv Example of 3rd order network edges

Use HONVis to visualize and interactively explore the higher-order network of NYC taxi data

HoNVis framework Given the networks, we provide three levels of exploration, including the global level to identify nodes of interest, individual level, and local level

HoNVis interface Design the interface based on five different views

HONVis: how to run How to run: Video tutorial of case studies: Windows: run HONVis_NYC_Taxi_Case_Jul_Aug.exe Mac and Linux: run ./Jul_Aug in Terminal Video tutorial of case studies:   Please follow the steps starting from 4:52 in the video HONVis_KDD2018.mp4 Please also refer to Section 7 in the attached paper for more explanation of the discoveries. Also refer to website http://www.higherordernetwork.com/

Appendix

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories

Synthesizing trajectories By t=1099, we have 200 first-order rules, 20 second-order rules, 12 third-order rules