Genii Weblog

Data mining thought for the day

Wed 16 Dec 2015, 09:24 AM



by Ben Langhinrichs
In between dealing with the horror that is Internet Explorer 9 and releasing new versions, I am working on a presentation on data mining in Notes rich text. With that in mind, here is my data mining thought for the day: 
 
There is implicit as well as explicit data and meta data. Explicit is there to be read, implicit is there to be discerned.
 
  1. Explicit data is the content (e.g., field in document; audio of phone call).
  2. Explicit meta data is the context (e.g., db and views where document is found; identity of callers and time of call).
  3. Implicit data is the internal implied context (e.g., words appear to be in English; caller sounds angry and agitated).
  4. Implicit meta data is the external cumulative context (e.g., occasional words in documents by this author appear to be German words which might imply native tongue; calls between person A and place of employment tend to be more agitated and frequent very late on Fridays which might imply somebody has to work weekends and is angry about it).
 
OK, back to Internet Explorer. If I am not heard from soon, send the Saint Bernard and brandy.
 

Copyright 2015 Genii Software Ltd.

What has been said:

No documents found