[Corpora-List] Open Research Position (M.S. / Ph.D. / post-doc):, Analyzing Routine Activities for Crime Prediction

[Matthew Gerber]

On Sun, May 25, 2014 at 5:57 PM, Christina Murmann
<christinamurmann at web.de>wrote:

>
>
> Matthew Gerber wrote:
> "I've provided victim-based examples"
> Even if we do not take a potential criminal point of view but a victim
> centered one: It is still questionable if the potential (!) victim's data
> should be used in such a way. After all, they did not ask for their
> whereabouts and daily routine activities to be monitored and used. And they
> will not be made aware of it either, I assume.
>

You have assumed incorrectly. All of our data are collected from Twitter's
public API. Specifically, we use the streaming API, which provides access
to Tweets that have been tagged with GPS coordinates. Users must opt in to
the use of GPS tagging. So, yes, these users have _specifically_ asked for
their whereabouts to be made public.

To the point of whether we should be using the data as we are, I can only
say that every single Twitter user in the history of Twitter has agreed to
Twitter's Terms of Use, which permit us to collect and use their data as
described. If people don't want us to use their Twitter data in our crime
research, they should not use Twitter. Nobody is forcing them to do so.


>  Also, the approach cannot be purely victim-based for obvious reasons: in
> two cities there might be blocks with similarly high proportions of 9 to 5
> workers and therefore - according to this approach - an alledgedly high
> risk of robbery. Still, past criminality rates and mean wealth might differ
> greatly for these cities, probably resulting in a quite different risk of
> robbery for the respective city. A purely victim based approach doesn't
> accout for that.
>

Certainly, it does not. That's why we do, in fact, look at these other
factors when we build our crime analysis
models<http://ptl.sys.virginia.edu/ptl/node/148>.
I was just trying to balance out the discussion's exclusive focus on
criminals.


>  Matthew Gerber wrote:
> "Now of course, all of this could be used for malicious purposes, just
> like every other technology in the history of Man."
>
> Matthew Gerber seems to want to say 'Because other technologies have had
> this flaw, I should not need to be troubled by this fact'
>

Not quite. I'm just saying there's nothing new here with regard to human
rights. People need to be aware of the content they post publicly and the
rights that other people (e.g., myself) have to collect and use such
information. This is an old, but worthwhile, discussion.

Sincerely,

Matthew Gerber
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20140526/4b5dd48f/attachment.htm>
-------------- next part --------------
_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora