An introduction to the machine learning library sklearn
which comes with Python.
The example in this project uses the supervised learning naive bayes algorithm.
The dummy data represents membership data and has a flag if these members purchased a hero product.
We have set two patterns in this data which is 35% of members who have purchased this product have a salary over £70,1234 and 25% of these members recently bought product 2024.
70% of the data is used to create the ML model. Then model is tested with the remaining 30%.
The accuracy should be 70% or greater.
The remaining 30% of data was tested and below are the results:
True Negatives: 21
True Positive: 6
False Positives: 0
False Negative: 4