Meeting Notes 20090717 NEMO Teleconference

From Nemo

Jump to: navigation, search

Attendees: Perfetti, Dou, Sydes, Connolly, Kilborn, LePendu, Frank, Curran, Frishkoff, Malony. [Dien not present.]

9:39 - start meeting, overviewed agenda (Frishkoff)

9:42 - introductions of "core" personnel (Frishkoff)

9:45 - funding awards are on the way and/or in-progress (Dou)

9:50 - introductions of consortium partners (Frishkoff), please send links to websites to Jason Sydes

9:55 - Round Robin introductions:

Connolly - ERP, MEG, fMRI work, speech processing, reading, homophone study, traumatic brain injury, greater Toronto area

Curran - ERP, memory, linguistic student doing metaphor comprehension, word studies

Kilborn - U Glasgow (OR native), newcomer to ERP (8 years), language (ephasia, bilingualism, aging), dementia assessments, industrial funding sources (pharms, etc), functional markers of disease show promise

Dennis ? - learning disabilities, sleep / sleeploss, school performance, (...)

Perfetti - U Pitt, word comprehension, N400, other markers, reading skills and learning ERP markers, co-work with Frishkoff, word recognition, spelling

10:03 - external collaborator introductions (Frishkoff)

10:07 - regular meetings using, once monthly (Frishkoff)

10:09 - project aims overview (Frishkoff); TODAY - ontologies as conceptualization or classification; NEXT TIME - analysis tools; FINALLY - database and portal (Frishkoff)

10:14 - NEMO website as central place to find resources -- sourceforge for ontology storage; NCBO tools for online browsing; sourceforge listserves for discussion; etc. (Frishkoff)

10:18 - "problem definition" (Frishkoff); ERP pattern labels mismatch, like apples and oranges or broccoli (Frishkoff)

10:21 - "why ontologies"; data + structure => information + semantics => knowledge; ontologies are very rich in semantics but it costs effort and time to build (Frishkoff)

10:25 - ontology design principles; OBO / NCBO best practices -- factor, reuse, validate, collaborate (Frishkoff)

10:29 - Factor -- spatial domain (e.g., topography) was our first stab due to its tractability (Frishkoff)

10:32 - Spatial Ontology overview, BFO, scap regions (ref: ICBO 2009 paper) (Frishkoff), relationships are key to reasoning power of ontologies (e.g., "rostral to"), SNAP BFO foundation

10:36 - Temporal Ontology preliminaries. Early, mid-latency, late classifications (Frishkoff), SPAN BFO foundation

Question: aside from coarse grained categories, can the ontology support more granular data? -- Yes, and more by reasoning across the semantics provided by the hierarchy, relationships, etc.

10:42 - Functional Ontology preliminaries. Brainmap as a model. CogPO. (Frishkoff)

10:45 - Pattern labels and rules defining them as the main goal for these orthogonal ontologies.

Question: (??) Given the pattern rules, how do the temporal notions fit in... starting to make more sense? Queries can operate a various levels of granlularity, over rules as well as hierarchy... (Perfetti and Frishkoff)

10:50 - Protege freeware allows you to interact with ontologies. (Frishkoff) NCBO provides similar online browsing tools.

10:53 - NEXT TIME: overview of tools for labeling data.

10:55 - Any questions?

Record meeting next time. Notes will be published and circulated online.

10:58 -- end of meeting.

/wiki/images/1/17/Fish1.png /wiki/images/e/ea/Fish2.png /wiki/images/f/fa/Fish3.png /wiki/images/f/ff/Fish4.png /wiki/images/4/40/Fish5.png /wiki/images/c/c5/Fish6.png
Personal tools