where to invest? - cloudinaryres.cloudinary.com/general-assembly-profiles/image/upload/v... ·...
TRANSCRIPT
Where to invest? Wanderson Batista Roldão
Data Science Workflow❖ Identify the problem
❖ Acquire the data
❖ Parse the data
❖ Mine the data
❖ Refine the data
❖ Build a model
❖ Present the results
Which is a good country to invest?Based on the Doing Business Indicator from the World Bank, specifically the costs of the % of the income per capita.
Data from official sources are based on different keys, but sometimes we don’t see what the data is trying to say.
Identify the problem
Aquire the datahttps://datacatalog.worldbank.org/dataset/doing-businessDoing Business website
Parse the dataOriginal Data (CSV)
Original Dataframe
Mine the data
Refine the dataTransform the data to a format that can be used for data science
Refine the dataMelt and Pivot the dataset
1. Melt
2. Pivot table
3. Rename columns
Build a modelThe proposed model is Unsupervised Learning: K-Means
Due to inconsistency with the resultsthe DataFrame was reduced to only the last year and removed NaN
Present the results
Silhouette Coefficient determinea number of cluster of 5
Present the results
Present the results
Present the results