" Semantics " for Advancement in Representation and Media: More quick witted Data Science

Semantics for innovation in visualization and multimedia smarter information science l.jpg
1 / 34
1369 days ago, 615 views
PowerPoint PPT Presentation
“ Semantics ” for Innovation in Visualization and Multimedia: Smarter Information Science. ICSTI Workshop February 8, 2011, Redmond WA. Diminish Fox (RPI) pfox@cs.rpi.edu Tetherless World Constellation http://tw.rpi.edu. If you don't mind clasp your safety belt. Working reason and the weight

Presentation Transcript

Slide 1

" Semantics " for Innovation in Visualization and Multimedia: Smarter Information Science ICSTI Workshop February 8, 2011, Redmond WA Peter Fox (RPI) pfox@cs.rpi.edu Tetherless World Constellation http://tw.rpi.edu

Slide 2

Please clasp your safety belt Working reason and the weight Opportunity – new means Linked open information (LOD) Open-source programming (Field) Science lead Semiotics of depiction Includes semantics Representing e.g. Vulnerability, quality, predisposition Speculation Tetherless World Constellation

Slide 3

Working reason Scientists – really ANYONE - ought to have the capacity to get to and utilize a worldwide, appropriated learning base of logical information that: has all the earmarks of being coordinated gives off an impression of being locally accessible But… information and data is acquired by different means (instruments, models, examination) utilizing different (frequently hazy) conventions, in varying vocabularies, utilizing (some of the time implicit) suppositions, with conflicting (or non-existent) meta-information. It might be conflicting, fragmented, developing, and disseminated AND made in a shape that encourages era, not use (with the exception of coincidentally) And … critical levels of semantic heterogeneity, expansive scale information, complex information sorts, legacy frameworks, unbendable and unsustainable execution innovation… Uh-gracious

Slide 4

Changing the condition " Changing the Equation for Scientific Data Visualization " – Fox and Hendler (Feb 11, 2011) Science (Perspectives), in press (banned, sad) Three essential focuses Unlocked information (and it " s enormous, ridiculously… ) Visualization for the masses all through the " life-cycle " however sans scale (!) Smarter information, more intelligent perception

Slide 5

.. Information has Lots of Audiences More Strategic Less Strategic From " Why EPO? " , a NASA inside write about science training, 2005 Science as well!

Slide 6

Shift the Burden from the User to the Provider – for Viz. as well! Fox Informatics and Semantics, © 2008

Slide 7

Too numerous outlines

Slide 8

Visualizing Linked Open Data (logd.tw.rpi.edu)

Slide 9

L inked open information Simply put: information is in RDF and has a URI and additionally it " s behind a question capable " triple-store " interface " change over " stack " inquiry " render "

Slide 10

New means – specialists to the safeguard Digital craftsmen, they required great inventive visual apparatuses, workmanship at the speed of innovative thought, feeling, instinct, mental representation and they adore programming And, RPI has EMPAC – Experimental Media and Performing Arts Center

Slide 11

From level screen to black box - EMPAC

Slide 12

Field – quick imagining

Slide 13

What we are doing Field meets Linux! Connected information meets Field! Nourish the current LOD illustrations into Field for control Then… . Unscrew the Google illustrations Unscrew the JSON sustain Query/devour crude RDF Visualizing at the speed of thought/writing.. From the tablet to scale So this is the place the semantics re-enter, particularly for depiction

Slide 14

L inked open information Field expends JSON (webify it) " change over " stack " inquiry " render "

Slide 15

L inked open representation Field questions " triple stores " (semantic webify it) " inquiry "/" render " change over " stack "

Slide 16

L inked open perception Field inquiries " triple stores " (semantic webify it) " question "/" render " dynamic " get to " stack "

Slide 17

Science - Means of direct

Slide 18

So shouldn't something be said about snatching? Actually no, not the criminal importance… Is a technique for intelligent derivation presented by Peirce which comes preceding acceptance and finding for which the conversational name is to have a "hunch". Abductive thinking begins when an inquirer considers of an arrangement of apparently disconnected actualities, outfitted with an instinct that they are some way or another associated goodness, hold up, great job for representation!!! Influence open world, semantics, as well… on the web

Slide 19

Information hypothesis Semiotics, likewise called semiotic studies or semiology, is the investigation of sign procedures (semiosis), or implication and correspondence, signs and images, into three branches: Syntactics: Relation of signs to each other in formal structures Semantics: Relation amongst signs and the things to which they allude; their denotata Pragmatics: Relation of signs to their effects on the individuals who utilize them

Slide 20

Semiotic model

Slide 21

Semiotics of depiction But we are discussing a computerized world progressively more than a simple one Beyond the detachment of substance from presentation We have implies for substance (setting and structure) semantics however pragmatics? Depiction (not simply " maps " or " charts " ) How – representation of substance, setting and structure, catch perception provenance Graphs, focuses, lines polygons, titles, tomahawks, shading, shade, measurements, … and their connection to each other!

Slide 22

For science viz. We leave the untold things untold – enormous (outrageously huge) problem(s) – like: Uncertainty Quality Bias Need prove A case?

Slide 23

MODIS Terra & Aqua versus Show Cloud Top Pressure AIRS versus MODIS Terra AIRS versus MODIS Aqua Correlation maps for Jan 1 – 16, 2008 Impact: Throw your hands uncertain and simply leave, quietly… MODIS Aqua versus MODIS Terra

Slide 24

Known Issues: The distinction of EQCT and Day Time Node, tweaked by information day definition, brought on the included bridge time contrast, which has the curio effect. See test pictures: BUT WHY ARE WE SAYING THIS IN WORDS? Included Overpass time Difference MODIS Terra versus MODIS Aqua AOD Correlation

Slide 25

Abductive Information System ? What might this look like in application apparatuses? How to investigate " hunches " (insights)? On the off chance that you assent that enlistment is in a general sense some portion of how a data framework is produced, then how to take into consideration snatching before acceptance might be conceivable? Open world, integrative Design elements? Engineering components? Library elements? Psychological elements?

Slide 26

Speculation But back to huge information and the need to transform the representation " dividers " into shows, 4-measurements – establishments – i.e. not submersion but rather encounter Synesthesia – why one and only sense? Quick

Slide 27

Speculation At scale – why? Stereo – why? Connected to the live information – insignificant curation! Objective: reestablish abductive thinking to the lead of science for experts and non-authorities And this must be informatics-based not some impromptu nerds making stuff up… Collaboration – wanna play?

Slide 28

So long and … pfox@cs.rpi.edu http://tw.rpi.edu http://openendedgroup.com http://logd.tw.rpi.edu http://empac.rpi.edu

Slide 29

Back shed

Slide 30

Need to be here Curation stages 20080602 Fox VSTO et al.

Slide 31

Mind the Gap! There is/was still a hole amongst science and the hidden framework and innovation that is accessible Informatics - data science incorporates the investigation of (information and) data, the act of data handling, and the designing of data frameworks. Informatics concentrates on the structure, conduct, and collaborations of common and manufactured frameworks that store, prepare and convey (information and) data. It likewise builds up its own particular reasonable and hypothetical establishments. Since PCs, people and associations all procedure data, informatics has computational, psychological and social perspectives, including investigation of the social effect of data advancements. Wikipedia. Cyberinfrastructure is the new research environment(s) that bolster propelled information obtaining, information stockpiling, information administration, information coordination, information mining, information perception and other figuring and data handling administrations over the Internet.

Slide 32

Modern informatics empowers another scale-free** structure approach Use cases necessities Stakeholders Distributed power Access control Ontologies Maintaining Identity

Slide 33

Multi-layered interoperability utilized by

Slide 34

Tetherless World Constellation tw.rpi.edu Future Web Science Policy Social Hendler Themes Xinformatics Data Science Semantic eScience Data Frameworks Fox McGuinness Semantic Foundations Knowledge Provenance Ontology Engineering Environments Inference, Trust Multiple depts/schools/programs ~ 35 (Post-doc, Staff, Grad, Ugrad)