This is where our team at Deloitte Stored our data and ipython notebooks for the Deloitte Microstrategy Data Blitz. While it does seem a bit roundabout to use Python for our data preparation, we wanted to use the tool we felt most confident in for natural language processing and manipulation.
Our client need that we addressed was understanding how the World Bank (essentially the IMF) could deploy capital directly to entrepreneurs to use their deployable capital in a better way similar to Kiva.
All of the data comes from Kiva, a non-profit involved in granting loans to recipients in developing economies. This data was found on Kaggle in a data for good repository.
The analysis is split into two parts. The first part is understanding where the loans were being used for by pulling the noun phrases from the loan statements. The second part was cleaning the data of outliers for each set of country data.