IW Meeting 2012-03-01

From Inference Web

Jump to: navigation, search

Contents

Meeting Information

Agenda

  • Meeting time to 3:00 +1 Cynthia, Jim,  Tim, Deborah
  • Around the room
  • Outstanding Actions
  • Focus: X prov-wg?

Attendees

  • Deborah (regrets)
  • James
  • Tim
  • Paulo (phone only)
  • Cynthia
  • Jim

Meeting Preparation

Around the room

(fill this out before meeting)

Cynthia
  • Updated latest fuse trace with iwbrowser view (this DOES NOT HAVE hasRawString - this is the latest):

  http://inference-web.org/iwbrowser/NodeSetBrowser?w=900&mg=5&st=Dag&fm=Raw&url=https%3A%2F%2Ftw1.tw.rpi.edu%2Ffuse%2FPracticalApplicationExampleTest%2Fpractical-application-hypothesis-justification.owl%23PracticalApplicationHypothesis

  • Previous fuse trace is also available to view from webdav (this HAS hasRawString):

  http://inference-web.org/iwbrowser/NodeSetBrowser?w=900&mg=999&st=Dag&fm=Raw&url=https%3A%2F%2Ftw1.tw.rpi.edu%2Ffuse%2Fhypothesis%2Femergence_hypothesis_2.owl%23Hyp99_NodeSet

James
  • Developed datacube encodings for NYU's feature data.
  • Currently working with D3.JS to generate a corresponding visualization of this feature data.
  • Developing an architectural modeling of hypothesis/indicator/feature lineage.
    • Objective: establish level of drilldown to capture for lineage
    • This will largely need to be established based on clearly defined use cases.
  • Developing a sample reasoning trace highlighting hypothesis deconfliction in CADRE's reasoning. 
Tim
prov-wg

http://www.w3.org/2011/prov/wiki/Meetings:Telecon2012.03.01#PROV-O.27s_prov-wg_telecon_report_2012-03-01

  • improving process of OWL updates
    • including as explicit ISSUES on tracker
    • tagging email subjects with [owl change]
    • using hg branching for the issue being addressed
    • referencing hg changeset wherever possible/appropriate
    • associating to ISSUE-265 when the OWL requires contortions, notes on how to avoid at http://www.w3.org/2011/prov/wiki/PIL_OWL_Ontology#RL_compliance
    • Stian making RL checker jar to ease ISSUE-265
  • Corral implicit issues into ISSUES (by Monday)
    • Tim and Mike working on exposing implicit
    • Mike, Tim, and Daniel working on organizing existing issues.
    • If you have a burning issues, please point it out to us.
provenance-based named graph query

Trying to combine https://github.com/timrdf/prov-lodspeakr/blob/master/components/services/provrdf-owl-coverage/queries/prov_select_latest.query and https://github.com/timrdf/prov-lodspeakr/blob/master/components/services/provrdf-owl-coverage/queries/from_provrdf.query to get

https://github.com/timrdf/prov-lodspeakr/blob/master/components/services/provrdf-owl-coverage/queries/prov_select_latest_objective.query

Outstanding action items

DONE REMAINS Jim McC todo - send email with the listing of prov terms integrated and how those map to pml 

ONGOING James  - please make a list of questions to ask our team if we  think - a) we can get - b) what we want to obtain - 

ONGOING c) discussion w/ Tim on encoding indicator/feature data in datacube  format - will allow us to leverage existing visualization technologies. 

DONE James - initial example by Feb 28  

TODO REMAINS for james next week - summarize status of what we have gotten or hope to get from NYU 

DONE - tim and james come up with an email suggesting that bae use the data cube vocabulary

TODO REMAINS - cynthia/paulo - schedule a call with paulo, deborah, cynthia on the  issues jiao brought (cc inference web) also send an email first with the  questions and cynthia's provisional answers to inference-web 

DONE : tim to propose changing time to 3pm on thursdays. ok with deborah, james, and cynthia 

TODO REMAIN: Tim to email paulo to ask. 

  • google docs?

Today's discussions

Around the room

Cynthia 

cynthia: their examples not have pml:hasRawString, so we can't see anything that we usually see.

RETURNED TO:  use of pmlj:hasConfidenceValue, any examples?

Paulo

been brainstorming on how to support climate science. Discussed with Jim Hendler yesterday with DOE and gov data. DOE does have good open data.

do not have good involvement on DOE data related to science.

TODO: Tim to give Ryan's contact info to Paulo.

RETURNED TO: PML API on github


Jim

RETURNED TO: mappings page.

James

data cube encodings with NYU. using to rep statistical data for FUSE. to calculate emergence of indicators

d3.js to make vis of NYU. getting up to speed on d3.js

TODO: Tim to review James' examples.

Working on architecture of lineages in FUSE (trace from hypoth to community domain machine learning routines in document corupus)

clarifying use cases

RETURNED TO: D3.js with Jim

Tim

see notes above.

no return tos.

Outstanding TODOs 

 change meeting time

RESOLVED meeting time change dto 3pm by consensus.

outstanding items

TODO: Tim to review James' examples data cube.

 Cynthia, jiao's problem, paulo

paulo needs to be reminded of these TODOs

TODO: Cynthia ask Patrick about twc local titanpad.

RETURN TOs

Paulo - pmlapi on github

paulo: can cynthia use?

cynthia: it is on our SVN. https://escience.rpi.edu/svn/public/pml-api/trunk/

no need to vpn to it.

https://scm.escience.rpi.edu/svn/public/pml-api/

Paulo wants BSD, it is not viral. Does not want it to be GPL.

Jim: not much diff between apache and BSD. Apache is BSD style. Diff might be aptetns

TODO: Tim to send pointers to License decision makers.

use of pmlj:hasConfidenceValue, any examples?

pmlj:hasConfidenceValue,

http://inference-web.org/wiki/Review_of_prov-xg%27s_Provenance_Vocabulary_Mappings

http://inference-web.org/w/images/b/b6/Pml-provenance.owl.owl.manual.graffle.pdf

this is a common property and can be used to describe an agent's confidence    on a piece of information or an action step.It is typically used together with an object property    indicating who asserted the confidence. For example, an inference engine may be one example.    Although we did not specified its cardinality, it is typically used only once in one individual.

TODO: model the confirming/disconfirming examples using ths property.

mappings page

http://code.google.com/p/magetab2rdf/wiki/Mappings

used the binary PROV properties in conversion.

implicit subclasses of Entity, Activity, Plan.

lebot@gemini:~$  curl -H "Content-Type: text/turtle" -d "<http://code.google.com/p/magetab2rdf/wiki/Mappings> a <http://xmlns.com/foaf/0.1/Document> ."
 http://sparql.tw.rpi.edu/services/datafaqs/util/wikitable-gspo
@prefix dcterms: <http://purl.org/dc/terms/> .

<http://code.google.com/p/magetab2rdf/wiki/Mappings> 
  a <http://purl.org/twc/vocab/datafaqs#Satisfactory>;
    dcterms:subject <http://www.w3.org/ns/prov-o/Activity>,
        <http://www.w3.org/ns/prov-o/Agent>,
        <http://www.w3.org/ns/prov-o/Entity>,
        <http://www.w3.org/ns/prov-o/Plan> .

<http://www.w3.org/ns/prov-o/Activity> a <http://example.org/ns#URI> .
<http://www.w3.org/ns/prov-o/Agent> a <http://example.org/ns#URI> .
<http://www.w3.org/ns/prov-o/Entity> a <http://example.org/ns#URI> .
<http://www.w3.org/ns/prov-o/Plan> a <http://example.org/ns#URI> .

TODO: Jim add an example following http://www.w3.org/2011/prov/wiki/PROV_examples

James D3.js with Jim

Jim, did you push your slides onto github?

on dropbox: http://dl.dropbox.com/u/9752413/d3talk/index.html

Datacube with d3: http://orion.tw.rpi.edu/~jimmccusker/datacube/


TODO: Tim, if the major predicate is in the row, turn that itn't a predicate. TODO: Jim to post issue on github

Facts about IW Meeting 2012-03-01RDF feed
Date1 March 2012  +
Personal tools
Navigation