2010Dec14

From Filtered Push Wiki
Jump to: navigation, search



User: David Lowery | User: James Macklin | User: Bob Morris | User: Zhimin Wang | User: BertramLudaescher | User: Lei Dou | User:
"User:" cannot be used as a page name in this wiki.
| User:
"User:" cannot be used as a page name in this wiki.
| User:
"User:" cannot be used as a page name in this wiki.
| User:
"User:" cannot be used as a page name in this wiki.
| User:
"User:" cannot be used as a page name in this wiki.



http://firuta.huh.harvard.edu:9000/cjwB1Kelp4

Report

  1. Zhimin: finished exposing GBIF data as rdf triples; working on interface design for annotations
  2. UCDavis(Lei and Bertram):Working on Curation package, finish general purpose Clustering and MailSender actor.
    1. The clustering actor clusters the list of record data on specified fields with specified clustering function.
    2. The MailSender actor authenticates user based on either username/password or OAuth token/secret and can synthesize the mail content by replacing the variables in specified template with incoming data.
  3. Bob: Been working exclusively on GBIF KOS report, but have learned a lot of new tools and possible FP usecase, e.g. circulating annotations from ontology repos such as BioPortal

Discussion

Bertram: ocurrence data (present and absence) are ided. Different concepts have some logical connections. Automatic alignment. Provenance story

James: Possible analysis workflow: 2 in proposal: georeferencing; number range. He will put some into the scenario repo in the wiki.

Bob: relevant fields partly missing

Bertram: do something real for key targets, which need not fansy and complete, to moblize user community.

Bob: organize around specify

James: has a good idea about the target usr: mainly natrual histroy. Demo over gbif data is a good start point.

Betram: touch clients as soon as possbile to undrstand their requiements

James: organize a workshop , may right after spnch meeting(May or June)

bob: dimension reduction: 50 attribute over millions of records. check assumption of relation between attributes. Constant quality checking like medacine validation, where with time going, you get more accurrate evaluation.