Three Steps for accurate ML Models
The Importance of Ratios & KPIs in Data Science
How to create useful Attributes in Machine Learning and co.
In my career, I came across the topics of building good ratios for Data Science and Statistics. These are essential for later good machine learning models and or can significantly improve this performance. But especially in the field of Business Intelligence this skill is also very important — because the formation of KPIs works the same way.
Building good Ratios in Data Science
To illustrate how meaningful ratios positively impact the outcome of a statistical model or machine learning algorithm in the best way possible, I would give an example using height, weight, and diabetes/glucose. For that, if you want to recreate it, you can download the data from this link [1]. Everything else I have coded and plotted with Jupyter Notebook.
The topic is, as I said, diabetes or, alternatively, the glucose value of a person and the relationship between height, weight and the ratio of the latter the BMI (Body Mass Index).
Step 1: Get the data from the CSV and filter it on certain columns:
import pandas as pd
#Read the csv
df =…
