Mike,
I know a couple of people trying to work on a similar problem: http://fictometer.com
They are using django as well, iirc.
Just a plug to what I came across in other places! Also, the problem is far from solved and they are trying some NLP approaches. That means good traditional way to prove things one way or the other is still open.
Abishek
On Mon, Sep 30, 2019 at 4:02 PM Mike Dewhirst <miked@dewhirst.com.au> wrote:
On 30/09/2019 4:56 pm, Mahir Shah wrote:
> I want to build an big project in django. Please can anyone suggest me some good projects based on the current world problems
The two biggest world problems I can see is group-think echo chambers
and fake news.
Such echo chambers probably come about from social media (Twitter,
Facebook etc) recommendations. Different people self-select their
favourite echoes. In the end, everyone seems to generate sufficient
meta-data to be slotted into particular categories as demonstrated in
2016 by Cambridge Analytica.
Although those categories become more and more granular as more and more
data accumulates for each individual that categorisation probably won't
go much further because the law of diminishing returns says it would be
too expensive to craft ever more targeted messages.
While such data is valuable for selling targeted advertising the big
problem I'm referring to is acquisition of such data by people or
organisations who cannot or should not be trusted with it.
The world needs a defence against stored meta-data. Not sure what you
might be able to do about that with Django but you did ask about current
world problems.
The second problem is, I think, a candidate for a Django project.
A lot of fake news is fed only into receptive echo chambers and
deliberately not fed into predictably skeptical echo chambers.
Together, these two world problems mean that unscrupulous people or
organisations can manipulate world events beyond the control of
democracies.
Fake news relies on assumed credibility. So what the world needs is a
news website which includes credibility links. It might work a little
like Wikipedia in that the news items posted might be curated by the
public in a couple of streams - for and against. Obviously, people who
uncover fake news would gain credits. People who support fake news would
lose credits.
Who knows? Maybe it would work. Maybe not.
Just in case, I have registered credipedia.com and if you come up with
something good I might just donate it to the cause :)
Cheers
Mike
> I am comfortable with any domain .
>
--
You received this message because you are subscribed to the Google Groups "Django users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-users+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/django-users/64c4c132-e1c1-70db-f1d8-77b0b73440c2%40dewhirst.com.au.
You received this message because you are subscribed to the Google Groups "Django users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-users+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/django-users/CAPm8%3Dxop4vNitm6g7WOUFyfM_WcrnAT-EOco6HRtTMb%3DJ6jRrg%40mail.gmail.com.
No comments:
Post a Comment