Thursday, August 7, 2014

What to do with GDELT

Here is why having GDELT accessible through Google cloud is a very cool thing: This dataset, containing 200 million world events, can be mined for analysis and prediction. Now, let's say I have a friend who has a method for profiling, and some knowledge about statistics, data mining, so now this friend can write this cloud based application that simply accesses GDELT data and offers services based on this data which would also be housed on Google Cloud (I am not saying right away -one day- I have a full-time job). The continuous update of the GDELT data would be taken care by Google. More important, access to data, which is huge, is handled through Bigquery that is optimized for big datasets such as GDELT. Actually curating / providing access to dsets like this is the perfect cloud centric service. It is the new age library service of sorts.

Note: I stumbled upon this dataset while I was looking for the world cup dataset (I was trying to export this data through my GC acct and saw gdelt) used by Google engineers for their world cup prediction. Last I checked their success was 8 in 8; remarkable.

Note: GDELT on GC is already being used for academic research. 

Q&A - 19/6

Bank of England The vast majority of money held by the public takes the form of bank deposits. But where the stock of bank deposits com...