Patrik, (01)
One major issue with Big Data (and any data, for that matter) is the issue of
the scope of the data sets, which is usually left implicit and often inferred
from external knowledge about the data source. By scope of the data sets I mean
what portion of reality does the data set purport to represent. Sometimes the
complexity of the data representation in a data set is due to explicit
inclusion of scope information, but usually scope is left unspecified in the
data representation. (02)
For example, if one is trying to determine air traffic patterns from data sets
provided by the various national/regional air traffic authorities or airlines,
aside from all the differences in representation and complexity of such data by
the different sources, one has to determine what portion of the overall air
traffic is captured by the aggregate sources one has access to, and whether
there is any overlap among the sources (and what the nature of the overlap
might signify with respect to one's objective for accessing the data sources).
Do some of the sources include general aviation traffic or only scheduled
commercial (passenger?) traffic. What portion of the world's air traffic (of a
particular set of types) do we not have data sources for? Are the time ranges
of the data sources compatible with the data access objectives? Does a
particular source include military aircraft traffic? Does it include charters.
Does it include Government executive aircraft? (03)
What about helicopter traffic or lighter than air traffic or UAVs? Up and down
to what vehicle size ranges? What about sub-orbital or orbital traffic (even if
one excludes "space" traffic as not being "air" traffic, space and orbital
traffic typically traverses the atmosphere when launched and often returns
through the atmosphere)? Are hovercraft considered air traffic? What about
gliders, paragliders, and "airsuits", or are we only interested in powered
aircraft or fuel-burning aircraft (not all powered craft burn fuel)? Note that
there are fuel-burning paragliders. Are rockets/missiles and artillery
considered "air" traffic? (04)
When one accesses Big Data for some purpose, what has one really accessed? How
big is "Big"? More importantly, how big a portion of what one is looking for
does Big Data represent? And what can one safely conclude for the purposes at
hand, given that scope information (assuming it is available or can be
inferred)? I'm not sure this is totally a question of logic. (05)
Hans Polzer (06)
-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Patrik Eklund
Sent: Friday, August 29, 2014 2:42 PM
To: ontolog-forum@xxxxxxxxxxxxxxxx
Subject: Re: [ontolog-forum] Looking to the Future of Data Science -
NYTimes.com - 2014.08.27 (07)
On 2014-08-29 20:47, Peter Yim wrote:
> ...
> http://bits.blogs.nytimes.com/2014/08/27/looking-to-the-future-of-data
> -science/ (08)
> ... The “big” in big data tends to get all the attention, Mr. Etzioni
> said, ... involves both grammar and background knowledge.
> And the latter is something humans acquire through experience of the
> world. (09)
Being new to this forum I may hit an "out of bounds", but my I7 tells me "big
data" should not even be about that "big". Size doesn't matter.
It's the complexity of data, and it's not even that. It's the complexity of the
structure of the data, like it's all about the complexity of the anatomy of the
swing, not just the complexity of the swing. A "big swing" mostly brings you
out of bounds, in particular when you focus on that "big". Or at least, it
produces a huge slice. (010)
Data "mining" is really silly, if you ask me, and I for one think we should
reserve "mining" just for minerals, not for knowledge. (011)
It's all about what logic really is, and isn't, isn't it? (012)
Cheers, (013)
Patrik (014)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/ Community Wiki:
http://ontolog.cim3.net/wiki/ To join:
http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (015)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (016)
|