AI/ML Experiments - Predictive Model Engineering
The Art Of Prediction
Machine Learning Experiments
One of the superpowers of AI/ML is the ability to predict future outcomes by training a model on historical data. Applications of such models are vast in business and creative contexts. I demonstrate two applications of AI/ML predictive engineering, below.
Churn prediction app
This machine learning model predicts if a customer will leave a particular bank or not. It's trained on the bank's public dataset of 8000 rows, with parameters specific to the bank. The model has an accuracy of 80%. Fill out the form and press predict!
Time series forecast deep learning app
Imagine being able to remove the guesswork from sales forecasting. We used deep learning to train a model to do just that. Our model was trained on 7 years of monthly Norwegian car sales data.