SEASR Overview

15
SEASR National Center for Supercomputing Applications University of Illinois at Urbana-Champaign Loretta Auvil [email protected] The SEASR project and its Meandre infrastructure are sponsored by The Andrew W. Mellon Foundation

description

Presentation given introducting SEASR on Mar 31, 2009 for ICHASS to the faculty from the UIUC Department of African American Studies

Transcript of SEASR Overview

Page 1: SEASR Overview

SEASR

National Center for Supercomputing Applications!University of Illinois at Urbana-Champaign

Loretta Auvil [email protected]

The SEASR project and its Meandre infrastructure!are sponsored by The Andrew W. Mellon Foundation

Page 2: SEASR Overview

TheSEASRPicture

Page 3: SEASR Overview

SEASROverview

SEASRwill:•  helpscholarsaccessexis9nglargedatastoresmorereadily

•  providescholarswithenhanceddatasynthesisandqueryanalysis–  fromfocuseddataretrievalanddataintegra9on

–  tointelligenthuman‐computerinterac9onsforknowledgeaccess

–  toseman9cdataenrichment

–  toen9tyandrela9onshipdiscovery–  toknowledgediscoveryandhypothesisgenera9on

•  empowercollabora9onamongscholarsbyenhancingandinnova9ngvirtualresearchenvironments

Page 4: SEASR Overview

AQuickLookatSEASR

•  Addresses:– Challengesoftransforminginforma9onintoknowledge

– Construc9ngthesoGwarebridgestomovefromtheunstructuredandsemi‐structureddataworldtothestructureddataworld.

•  Aims:– Makedigitalcollec9onsmoreuseful– Provideaccesstorelevantanaly9csandvisualiza9ons

– EnableeasymashabilityviaSOA

Page 5: SEASR Overview

SEASR:Reach+Relevance+Reuse+Repeatability

SEASRemphasizesflexibility,scalability,modularity,providescommunityhubandaccesstoheterogeneousdataandcomputa9onalsystems–  Seman9cdrivenenvironmentforSOAinteroperability–  Encouragessharingandpar9cipa9onforbuildingcommuni9es–  Modularconstruc9onallowsflowstobemodifiedandconfiguredto

encouragereusabilitywithinandacrossdomains–  Enablesamashupandintegra9onoftools–  Data‐intensiveflowscanbeexecutedonasimpledesktoporalarge

cluster(s)withoutmodifica9on–  Computa9oncanbecreatedfordistributedexecu9ononserverswhere

thecontentlives–  Useraccessibilitytocontroltrustandcompliancewithrequiredcopyright

licenseofcontent–  ReliesonstandardizedResourceDescrip9onFramework(RDF)todefine

componentsandflow

Page 6: SEASR Overview

SEASRTextAnaly9csGoalsAddresstheScholarlytextanaly9csneedsby:

•  EfficientlymanagingdistributedLiteraryandHistoricaltextualassets•  Structuringextractedinforma9ontofacilitateknowledgediscovery•  Extractinforma9onfromtextatalevelofseman9c/func9onal

abstrac9onthatissufficientlyrichtosupportques9on‐answering•  Devisearepresenta9onfortheextractedinforma9onthatcanbe

efficientlyreasonedovertorecoverdataintheques9on‐answerprocess

•  Devisealgorithmsforques9onansweringandinference•  DevelopUIforeffec9vevisualknowledgediscoverywithseparate

querylogicfromapplica9onlogic•  Leveragingexis9ngapproachesanddevisealgorithmsforclustering,

inference,andQ&A•  DevelopinganInterac9onUIforeffec9vevisualdataexplora9on•  Enablethetextanaly9csthroughSEASRcomponents

Page 7: SEASR Overview

Workbench

•  Web‐basedUI

•  Componentsandflowsareretrievedfromserver

•  Addi9onalloca9onsofcomponentsandflowscanbeaddedtoserver

•  Createflowusingagraphicaldraganddropinterface

•  Changepropertyvalues•  Executetheflow

Page 8: SEASR Overview

CommunityHub

Page 9: SEASR Overview

SEASR@Work–Zotero

•  PlugintoFirefox•  Zoteromanagesthe

collec9on

•  LaunchSEASRAnaly9cs–  Cita9onAnalysisusestheJUNG

networkimportancealgorithmstoranktheauthorsinthecita9onnetworkthatisexportedasRDFdatafromZoterotoSEASR

–  ZoteroExporttoFedorathroughSEASR

–  SavesresultsfromSEASRAnaly9cstoaCollec9on

•  LaunchMONKProcessing–  MONKDBInges9onWorkflow

Page 10: SEASR Overview

WebService

Interac9veWebApplica9on

SEASR@Work–Fedora

Page 11: SEASR Overview

SEASR@Work–En9tyMash‐up

•  En9tyExtrac9onwithOpenNLP

•  Loca9onsviewedonGoogleMap

•  DatesviewedonSimileTimeline

Page 12: SEASR Overview

SEASR@Work–AudioAnalysis•  NEMA:ExecutesaSEASR

flowforeachrun

–  Loadsaudiodata–  Extractsfeaturesforevery

10secmovingwindowofaudio

–  Loadsandappliesthemodels

–  SendsresultsbacktotheWebUI

•  NESTER:Annota9onofAudioviaSpectralAnalysis

Page 13: SEASR Overview

SEASR@Work–MONK

Executesflowsforeachanalysisrequested– Predic9vemodelingusingNaïveBayes

– Predic9vemodelingusingSupportVectorMachines(SVM)

Page 14: SEASR Overview

SEASR@Work–DISCUS•  On‐demandusageof

analy9cswhilesurfing–  Whilenaviga9ng

requestanaly9cstobeperformedonpage

–  Textextrac9onandcleaning

•  Summariza9onandkeyworkextrac9on–  Listtheimportant

termsonthepagebeinganalyzed

–  Providerelevantshortsummaries

•  Visualmaps–  Provideavisual

representa9onofthekeyconcepts

–  Showthegraphofrela9onsbetweenconcepts

Page 15: SEASR Overview

SEASRandUIMA:Emo9onTrackingGoalistohavethistypeofVisualiza9ontotrackemo9onsacrossatextdocument(Leveragingflare.prefuse.org)