AI Data Science 24
Back Order Prediction
To build a model which will be able to predict whether an order for a given product can go on backorder or not. A back order is the order which could not be fulfilled by the company. Due to high demand of a product, the company was not able to keep up with the delivery of the order.
Architect
Model Training
1. Data Export from Db -
The data in a stored database is exported as a CSV file to be used for model training.
2. Data Preprocessing
Check for null values in the columns. If present, drop the null values.
Check if any column has zero standard deviation, remove such columns as they don't give any information during model training.
Apply scaling and PCA in the columns , remove multi collinearity.
3. Model Selection
After clusters are created, we find the best model for each cluster. We are using two algorithms, "Random Forest" and "XGBoost". For each cluster, both the algorithms are passed with the best parameters derived from GridSearch. We calculate the AUC scores for both models and select the model with the best score. Similarly, the model is selected for each cluster. All the models for every cluster are saved for use in prediction