eScience Group August 05, 2014

Printer-friendly version

General Meeting Information

  • Previous Meeting
  • Call-in information
    • Goto Meeting
      • Dial +1 (805) 309-0012
      • Access Code: 776-009-689
      • Audio PIN: Shown after joining the meeting
      • Meeting ID: 776-009-689

Presentations

  • Keep this list from week to week so we know who’s presented and who will present.
  • Please sign up here if you have a good topic to share with others.
  • May. 27, 2014 - introduction, summer plans, no presentation
  • Jun. 03 2014 - Linyun: programming the semantic web 
  • Jun. 10, 2014 - ODIP group (Cyndy, Bob, Adam S., Adam L., etc.) will be at TWC and use 1142 
  • Jun .17, 2014 - Jin: Entity Linking System Tutorial
  • Jun. 24, 2014 - No presentation
  • July. 01, 2014 - Marshall: Golden Spike information portal: a mash-up of vocabulary encoding, SPARQL query and data visualization for the domain knowledge of geologic time
  • July. 08, 2014 - No presentation
  • July. 15, 2014 - Paulo: Jefferson Project
  • July. 22 2014 - Linyun: provenance capturing methods
  • July. 29, 2014 - No presentation
  • Aug. 05, 2014 - Peter Fox: event calculus

Potential

  • James Michaelis present on provenance work
  • Jefferson Project 
  • ToolMatch - Matt Ferritto
  • Anirudh - real time streaming data and analytics (PNNL)
  • GeoData 2014 report - Corey/Marshall/Peter
  • ESIP report - Patrick/Stephan/Peter
  • Web Science 2014 report (is anyone attending escience meetings from there)
  • Fall meeting sometime Stephan to present on data.rpi.edu

Attendance

  • Patrick
  • Corey
  • Han
  • Linyun Fu
  • Matt Ferritto
  • Hao Zhong
  • anirudh prabhu
  • Lakshmi Chenicheri
  • Stephan Zednik
  • momo
  • Benno
  • Sumithra
  • Peter Fox
  • Deborah McGuinness
  • Jin

Agenda

  • what everyone has been doing during summer - projects and research
  • Prof. Peter Fox’s talk: event calculus

Past Action Items

  • None

Action Items


Notes - Peter's talk on event calculus

  • Use case: how a volcano eruption will influence the environment, property labels as physical processes. Make explicit descriptions of processes, and reason on them.
  • event calculus vs. situation calculus
  • event calculus is able to handle both continuous and discrete changes
  • additive effects in event calculus
  • spotlight: separation of logic and equations reasoning
  • the partValue predicate, and breaksXXX predicates
  • future work: durative events, delayed effects, abductive reasoning
    • process modeling for the Web and the related tools to support authoring, parsing, etc.
  • Scalability depends on mathematical calculation efficiency

Notes - Who's doing what during the summer

  • Corey: "Boundary" activity in DCO -- extract data from scanned version PDF files
    • Boundary activity meaning that we have individual projects with various science groups that are members of the Deep Carbon Observatory (DCO) community. They have specific needs that might be different from other groups. 
    • In one case there are scanned PDF documents, older datasets perhaps, that we hope to scan and convert to actual data. It has been difficult to find tools to read table data from a scanned PDF.
    • In another case there are graphics, such as plots, that we hope to be able to pull data from. There are tools that can do this.
    • Once the data is "scanned" it has to be visually inspected to verify accuracy and quality.
    • Link to write up http://tw.rpi.edu/web/project/DCO-DS/WorkingGroups/BoundaryActivities/DataSource4Properties
  • Lakshmi: Creating 5-6 visualizations with d3 for DCO, working on the final report right now.
    • Additional boundary activities for Deep Carbon
    • Generating final report for the summer, will have available at the end of the week
  • Benno Lee: RDESC - Resource Discovery for Extreme Scale Collaboration
    • Big Data ... we have 4.5 billion triples in the triple store
    • Attempting to verify that we can use the existing tools within the lab to be able to use and visualize all those triples
    • Using twsparql, s2s (doesn’t work very well with so many triples, looking for ways to optimize queries, using indexing, possibly relational tables for more common queries, caching, etc...
    • http://rdesc.org
  • Matt: ToolMatch tool-dataset matching, working on the service for that, search and match for 2 use cases.
  • Linyun: the great and fabulous
  • Han
  • Momo
  • Anirudh
    • PNNL - Analysis in Motion
    • Similar to toolmatch (http://toolmatch.esipfed.org), includes CMAP, endpoint, etc...
    • Presenting poster at annual meeting for Analysis in Motion
  • Patrick West
    • DCO - SSO with Yu, will put into production this week or next
    • ToolMatch with Matt
    • RDESC with Benno
    • CMSPV, ECOOP with Linyun
    • OpeNDAP provenance with James Michaelis - trace of data processing (de borah just came at the end of that - i will read the email and see if we are ready to open it up for comments/submission)
  • Stephan Zednik
    • New Lord and Master of S2S, where we’re going in the future with it
    • Next Generation Rensselaer Data Services (data.rpi.edu), similar to DCO architecture