Building your first Machine Learning Model - Predicting customer behaviour with KNN classification

This project is aimed towards beginners and involves building a KNN classification model using Scikit-Learn. The purpose is to predict whether a customer is likely to buy a product based on their past purchase behavior. The project will also cover feature engineering techniques such as one-hot encoding, feature selection, training and testing of the model. Additionally, it will demonstrate how to save models as pickles, load them in a new environment, and predict output for a new set of data. All steps will be explained in detail for a completely hands-on learning experience.
Github link for source code and data : github.com/fut...
Python crash course : • Python NumPy Pandas Ma...
FutureX Tech playlist - • Tech - Cloud, Data and AI

[@muhammadshahin3667]

Very nice. I must point out one thing, though.
You are supposed to fit_transform the training data only. Subsequent scaling should only transform new data using the scaler fit earlier on the training data. This is very important to prevent data leakage.

[@tsionsamuel5885]

Thanks for the very nice explanation.

[@svezakaavezaka5552]

Can you give me the link of the dataset? The origin of it (like kaggle)? Also, how to interprate the column price, when its not the same for one product in two purchases? What does the purchase column mean? Is it for the new product that we are observinf or for that product

[@futurexinsights]

Please find it here github.com/futurexskill/ml-model-deployment/blob/main/storepurchasedata.csv

[@shaahidhirfaansha78]

what if the dataset does not have purchased or not target variable?

[@futurexinsights]

that becomes unsupervised learning