SBU CSE 591 - Lecture 5 - Computational Modeling and Graph-Based Analytics

Unformatted text preview:

CSE 591: Visual AnalyticsLecture 5: Computational Modeling and Graph-Based AnalyticsKlaus MuellerComputer Science DepartmentStony Brook UniversityMotivating QuoteGary Ackerman(Director of the Center for Terrorism and Intelligence Studies)“There are tools where they build a world in a bottle. They put down every single mosque, river, camel, and school in, say, Saudi Arabia. Then they have millions of software agents who each have desires, grievances, all these different variables. They go about their little lives and then you ask a question: What if we build a McDonald’s in Mecca? Does this lead to more people joining terrorist groups or not?”Motivating QuoteGary Ackerman(Director of the Center for Terrorism and Intelligence Studies)“There are tools where they build a world in a bottle. They put down every single mosque, river, camel, and school in, say, Saudi Arabia. Then they have millions of software agents who each have desires, grievances, all these different variables. They go about their little lives and then you ask a question: What if we build a McDonald’s in Mecca? Does this lead to more people joining terrorist groups or not?”The key message here: build of model of the “enemy” and predict what they will do given some input and situation•bound to overcome many of issues that came up in the “Psychology of Analysts” bookSimulations via Game Technology: The SimsSimulations via Game Technology: Unreal Tournament Games Have Analytical ToolsExample: Trust graphBarry Silverman (U Penn):“A jihadist terrorist has a preferred state of the world: the whole world is fundamentalist, and no loose-valued Westerner should walk on the land that Muhammad felt was sacred,”Long-term preferences of the “jihadist terrorist”:•control of the land• control of who’s allowed to do whatMoral standards:•it’s okay to kill • in fact, I’ll go to heaven and be a martyr• and my family will get moneyMarc Sageman (U Penn)A typical jihadist terrorist:•Is a male who grows up in a decent family, never gets into much trouble. • Goes to a religious school. There he learns Islamic fundamentalist credos.• In many cases he goes abroad to study—often engineering and often in Europe, where he feels alienated and has trouble finding work.• He then gravitates to mosques to find comfort in the familiar, and there he meets men with similar feelings of isolation. • Gradually, he is drawn toward a politicized existence and into aterrorist cell. By incorporating more than a dozen such models and data sets, some of Silverman’s simulations can emulate how a jihadist agent might feel about his group, when he will become radicalized, and what he might do as a result.Getting Inside The Mind: Overall System (Silverman): Getting Inside The Mind: NarrationEach of the simulated terrorists responds to stimuli through a complex process that mimics human thinking. An agent’s perception is influenced by various physical and psychological stressors and coping styles. •this perception is then viewed in light of an agent’s values, emotions, and culture. Once the agent has processed its perception of the stimuli, it must decide how to best attain its goals. •It evaluates which actions will advance its values, but this is subject to constraints of its social relations, emotions, and stress. • When the agent determines what it wants to do, it expresses thatdecision through action.Graphs, Hierarchies, NetworksPowerful way to show relations:Kathleen Carley (Carnegie Mellon)compressed view of Iraq’s Diyala regionexpansion into known connections among suspected insurgents(right after US invasion):sized relates toimportanceReferenceH. Goldstein, “Modeling Terrorists”, IEEE Spectrum, pp. 26-34, 9/2006. News and Blog AnalysisLydia•system developed right here at Stony Brook• Mikhail Bautin, Anand Mallangada, Alex Turner, Lohit Vijayarenu, Steven Skiena• was presented at SBU-CS Graduate Research Day 2007Large-Scale News AnalysisLydia news analysis system does a daily analysis of over 1000+ online English and foreign-language newspapers, plus blogs, RSS feeds, and other news sources.Currently tracks over 1,000,000 news entities, providing spatial, temporal, relational and sentiment analysisThe data and analysis should be of great interest in political science and related fields.Discussed here: Processing of Foreign Language Content, Entity Relation Extraction and News Storm detection of LydiaTextmap.comRelationship IdentificationUses hand-crafted regular expression rules that are fast and efficient in processing gigabytes of newspaper data.Looks for several specialized classes of relations like relations between family members, employer-employee relations, political relations etc.,Regex rules enable to trap relations such as “Maurice is the brother of Curtis Jones, who is the grandfather of Karli Jones”Sentiment AnalysisSentiment analysis allows one to measure how positively/negatively an entity is regarded, not just how much it is talked aboutSentiment AnalysisSentiment Index of “Korea” for five different languagesSentiment AnalysisNews Storm for “Sen. Bill Nelson”Agenda For Remaining WeeksRemaining lectures (13 sessions):•somewhat follow the book “Illuminating the Path”• use references therein to identify papers to read and present (plus a few more)• 2-3 papers per presenter and session• pick a topic and correspond with me to identify suitable papers• topics will be announced next Tuesday in class and then on the websiteFinal project:•chosen presentation topic can relate to final project (but does not have to be)• suggest a final project shortly after picking the paper topic• communicate with me to define project more closelyWeek after next (Tuesday or Thursday): •guest lecture by Prof. IV Ramakrishnan• topic: machine reasoning and


View Full Document
Download Lecture 5 - Computational Modeling and Graph-Based Analytics
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Lecture 5 - Computational Modeling and Graph-Based Analytics and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lecture 5 - Computational Modeling and Graph-Based Analytics 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?