View source for Data Mining for Viability -- Junk Monkey

{{:Summer of Content project proposal header}}<!-- DO NOT TOUCH THIS LINE --> 

== Junk Monkey ==

OLPC's goal of providing computers to underdeveloped countries will be immensely supported by thin-client software. If users can have an internet signal, they can then access enormous amounts of information, tools and software informing users of worldwide events while aiding local, community impacts.


The rise of blogs, e-newsletters and online newspapers compliments the rise of thin-client software. Much of the news information we receive on the internet has very little obvious metadata, such as background about the author or story subject.


This has bothered me and I believe it presents a major challenge for the future of information, not least in the developing world where blogs and online newsletters will undoubtedly flourish as computing service become cheaper and more readily available. To address this problem, I'd like to program a plug-in for Firefox that mines body text of news websites and cross-references data against databases.


The proposed databases can be already existing (such as PubMed, Wikipedia or Sourcewatch) or stimulated by the formulation of such a plug-in (a database of all journalists and major blog writers, for example).


I believe the project will be very challenging theoretically, while not incredibly difficult technically. I've tested some basic algorithms by taking body text, creating a few rules then analyzing the text. The results have been very promising and I hope to explore this idea with any interested peeps.


=== Examples ===

Theoretical examples are [http://themanhattanproject.wikispaces.com/JunkMonkey here] on the Project's wiki.

== Interested interns ==

=== Intern name ===

Contact information, why you'd be good for the job, any specific plans, variants, or details you would personally like to implement and why

[[Summer of Content interns#Hemant Goyal|Hemant Goyal]]-
[http://wiki.laptop.org/go/User:Hemant_goyal contact information]

[[Summer of Content interns#Vijay Majumdar|Vijay Majumdar]]

== Interested mentors ==

=== Mentor name ===

Contact information, why you'd be good for the job, any specific plans, variants, or details you would personally like to implement and why

== Other interested parties ==

=== Coogan Brennan ===

coogan(dot)brennan(at)columbia(dot)edu -- project proposer