This project is an interactive part of a bigger Adults project, so you are very welcome to visit it on my website and try it on your own
- the Random Forest Classification model is used for training
- data from the user is collected and transformed into the format, required by the model
- model predicts the income class
- the graph with Shapley values (feature importance) is built based on the prediction
Here a new 'Country' file is used to let the user select among all countries in the world. The custom column 'developed' is added with true/false
values based on Table A of this United Nations country classification report. This column is needed to transform the data to feed to the model.
The following data is collected from the user to make a prediction:
- Personal info
- Age
- Sex
- Country
- Ethnic group
- Family info
- Marital Status
- Family Belonging
- Education and professional info
- Education level
- Amount of working hours per week
- Workclass
- Occupation
- Capital operations Info
- Capital gain
- Capital loss