Predictive Analysis of Massive Streaming Graphs

Slides:



Advertisements
Similar presentations
Summary XBRL Challenge Objective: Tools that rely on XBRL data, e.g., tool that extracts data for multi-company comparison via desktop application; or.
Advertisements

“High Performing Financial Institutions and the Keys to Success in an Uncertain Environment”
Nick Wainwright HP Labs / Effectsplus project. The report of a consultation of the Future Internet Assembly – a cross disciplinary assembly of researchers.
Massive Streaming Data Analytics: A Case Study with Clustering Coefficients David Ediger Karl Jiang Jason Riedy David A. Bader Georgia Institute of Technology.
Tyson Condie.
An Online Knowledge Base for Sustainable Military Facilities & Infrastructure Dr. Annie R. Pearce, Branch Head Sustainable Facilities & Infrastructure.
Software Project Management With Usage of Metrics Candaş BOZKURT - Tekin MENTEŞ Delta Aerospace May 21, 2004.
INTERACTIVE ANALYSIS OF COMPUTER CRIMES PRESENTED FOR CS-689 ON 10/12/2000 BY NAGAKALYANA ESKALA.
KS3 Phase4 Client Server Monitoring System October 1, 2008 by Stephen, Seema, Kam, Shpetim.
Computing and SE II Chapter 15: Software Process Management Er-Yu Ding Software Institute, NJU.
ICN Baseline Scenarios draft-pentikousis-icn-scenarios-04 K. Pentikousis (Ed.), B. Ohlman, D. Corujo, G. Boggia, G. Tyson, E. Davies, P. Mahadevan, S.
Adaptive Integrated Framework (AIF): a new methodology for managing impacts of multiple stressors in coastal ecosystems A bit more on AIF, project components.
Homeland Security UNCLASSIFIED Coast Guard Cyber Strategy Awareness Training.
Www3.informatik.uni-wuerzburg.de Institute of Computer Science Chair of Communication Networks Prof. Dr.-Ing. P. Tran-Gia Modeling User Activity and Network.
NYP/Q PPS PAC Meeting NYP/Q PPS September 22, 2015.
BAHIR DAR UNIVERSITY Institute of technology Faculty of Computing Department of information technology Msc program Distributed Database Article Review.
What we mean by Big Data and Advanced Analytics
Creating tools for the mobile campus
PREPARED BY G.VIJAYA KUMAR ASST.PROFESSOR
Kai Li, Allen D. Malony, Sameer Shende, Robert Bell
IOT – Firefighting Example
Systems integration and Testing INSE 6421
City of New Haven, CT VHB COMMUNITY ENGAGEMENT COMMITTEE MEETING:
The Five Secrets of Project Scheduling A PMO Approach
Professor Harry Scarbrough
Julia Lane New York University
Ralleo Enterprise-Grade Solution for Managing Change and Business Transformation Provides Opportunities to Better Analyze Real-Time Data MICROSOFT AZURE.
Caliper Event Integration Blackboard Learn & Kaltura Video Platform
Research Methods Dr. X.
Responding to Intrusions
Project Management Managing Project Execution
Formulate the Research Problem
California Cybersecurity Integration Center (Cal-CSIC)
World-Leading Research with Real-World Impact!
Software Engineering (CSI 321)
Alma Consortial Analytics
Wenjing Lou Complex Networks and Security Research (CNSR) Lab
Z26 Project Management The Effectiveness of Workshops Graham Collins University College London (UCL)
John D. McGregor Session 9 Testing Vocabulary
An Application in Analyzing Isolated Malicious Activities
Goal, Question, and Metrics
John D. McGregor Session 9 Testing Vocabulary
Prasenjit Ghosh. Director Balram Mishra. Project Manager
1Georgia Institute of Technology 2Chicago Booth School of Business
Principles of Project Management Case Study - Programme Management: Developing a Blueprint Graham Collins University College London (UCL)
Safety Culture Self-Assessment Methodology
Age Friendly Communities
Risk Detection, Operations Efficiency & Economic Analysis Eva K Lee
University of California—Irvine
TRAINING FOR THE NETWORKED BATTLEFIELD Alice F. Healy and Lyle E
Crypteron is a Developer-Friendly Data Breach Solution that Allows Organizations to Secure Applications on Microsoft Azure in Just Minutes MICROSOFT AZURE.
Carl Data Solutions Collects Utility Sensor and Meter Data to Provide Advanced Reporting, Alarming, and Analytics with Microsoft Azure MICROSOFT AZURE.
Resource 1. Evaluation Planning Template
Development of a Real-Time Decision Support System for Proactive Response, including Recourse Planning, under Uncertain Active Cascading Emergencies PI:
Strategic Planning Summary.
Use of Time-Series Data in Strategic Managerial Decisions
Sightings and Observations
Analytics for tracking student engagement
Faculty Peer Review of Teaching
Presentation to Project Certification Committee, DoIT August 24, 2008
Faculty Peer Review of Teaching
Social Value Research Proposal
Cyber Security: What the Head & Board Need to Know
OSSArcFlow Project Overview
Kostas Kolomvatsos, Christos Anagnostopoulos
KEY INITIATIVE Financial Data and Analytics
Sustaining Repositories
World-Leading Research with Real-World Impact!
Nome Sobrenome. Time time time time time time..
WORK STREAM TEAM DELIVERABLES
Presentation transcript:

Predictive Analysis of Massive Streaming Graphs David A. Bader (PI) Georgia Institute of Technology Atlanta, Georgia Problem Statement and Objective: Develop methods using streaming graph framework (STINGER) to support update of quantitatively predictive metrics Move beyond current methods investigating historical data and building models on what has already been seen Problem Focus 1: Provide unclassified pre-event security reports to local and state law enforcement based upon streaming analysis of open source and public data sets. Working with DHS, NYPD, and FBI, focus on public events (e.g. concerts, sporting events) and financial district of New York City. Problem Focus 2: Provide Coast Guard with additional cyber-analytic tools based upon streaming graph analysis for identifying, tracking, and giving attribution to, cyber-threats. Methodology and Data Requirements: Add capabilities to STINGER to be able to use graph-based analysis and quantitatively measured results across a range of applications Per-vertex graph metrics, machine learning, complex analytics (i.e. community membership) In Problem Focus 1, we plan that all data will be open source and publicly available (e.g. public Tweets) In Problem Focus 2, we plan that data will be sensitive network traffic and housed with Coast Guard. We will use synthetic and surrogate data sets for development and testing, and transition the cyber-analytic codes to the component. Impact Statement and Relevance to DHS Roles and Responsibilities: STINGER metrics will be able to predict events and correct for what actually occurs in a security context STINGER will support faster than real-time prediction with high-performance computing methods for real-world problems PI Bader plans to work with co-investigators Jason Riedy, Umit Catalyurek, and Polo Chau (Georgia Tech) and Megan Cream (Spelman College) Multiple Engagements with DHS components: Dr. Kay Mereish: DHS I&A, NYPD, FBI, DHS Nebraska Ave Compound LCDR Ryan Lamb: Coast Guard Laura Laybourn: DHS NPPD Timeline and Deliverables: 30 days: Initial task work plan submitted (by October 20) 60 days: Firm engagement with specific components outlined here 90 days: Report on specific engagement with components, articulation of problem, methodology and data requirements, and transition plans Year 1: Reports evaluating existing analytics and new methods using per-vertex metrics Year 2: Reports evaluating new methods for predictive modeling, documentation to show development and plication of these methods using STINGER