IW Meeting 2010-07-29

From Inference Web

Jump to: navigation, search


Meeting Info


  • Li Ding
  • Paulo Pinheiro da Silva
  • Cynthia Chang

IW Overall Objectives

     1) PML ontology: (OWL level only) potential non-monotonic  changes (Li) 
     2) PML practice  
         2.1) Use cases and examples (Tim) 
         2.2) RDF level issues (Cynthia) 
         2.3) PML Tools / API issues (Cynthia) 
         2.4) PML primer (Tim) 
     3) PML and OPM mapping (Paulo) 
     4) IW management (TBD)
        4.1) inference web infrastructure and research plan
        4.2) technologies for engagement 

Today's Agenda


Issue 1 (close)
Issue 2 (close)
  • tim has account on iwregistry, we need to know his experience
Issue 3 (old proofs)

(cynthia) wine proof status

  • the wine proofs i saw were from wino. they are all in pml1. they look like the pre-generated proofs for wine pairing for different food. the logical sentences are written in kif. i see there are two parts to get these proofs to be useful. one is to convert them to the pml2 and the other is to make the application. i am not sure the wine agent james worked on generates the same proofs.

(paulo) these are not wine proofs:

  • 1) proofs that demonstrate our capability of verifying PML. That was done by pryiandra and it has PPDR (or inference ML) stuff in it;
  • 2) proof involving typicality assumptions in support of NIMD;
  • 3) JTP proofs derived from IBM UIMA logic;
  • 4) combined proofs of JTP reasoning and UIMA-extracted knowlege; mississipi is just UIMA-extracted knowledge. We used to have examples of COMBINED proofs. proofs is 4 are very large (probably the largest proofs we had at that time)
  • 5) proofs involving the use of tactics and declarative rules (possible also with the help of Inference ML)
  • 6) proofs involving tasks such as proofs derived by SRI SPARK; I am talking about CALO and Alyssa may have access to some of these proofs


  • [11:32:12 AM] Paulo Pinheiro da Silva: most of the proofs above are in PML1.
  • [11:33:19 AM] Paulo Pinheiro da Silva: my understanding is that we need to recover the proofs above and probably convert some of them to pml2. The mississipi example is already 'recovered'.
  • [11:33:58 AM] Paulo Pinheiro da Silva: not only convert, but move them to a more stable place and to have a proper place where we list them as demos
  • [11:34:38 AM] Paulo Pinheiro da Silva: sure. I agree with your approach for demo them.
  • [11:39:14 AM] Paulo Pinheiro da Silva: for (1), pryiandra; for (2), (3) and (4) , Bill Millar, for (5) Dhyanish and vasco; for (6) Alyssa


  • issue 1 (upload paper) closed
  • issue 2 (unzip rule) wait for Tim's experience
  • issue 3 (old proofs) paulo want more
  • ACTION: decide whether convert wine to PML2 next week
  • ACTION: cytnhia (with help from paulo and li) find proofs meeting the six categories listed by paulo
Personal tools