DATA SCIENCE - Urban Delivery Demand Prediction 100%

[blessy05]

This PR includes preprocessing and feature engineering for the GraphSAGE model.

Changes made:

• Cleaned dataset (handled missing values and duplicates)
• Converted time columns to datetime format
• Created new features:
  • order_hour
  • order_day
  • order_weekday
  • is_weekend
  • delivery_duration_mins
• Prepared dataset for graph-based modelling
• Implemented initial GraphSAGE model notebook

This forms the foundation for further model training and evaluation.

[nadunhs25]

Great work! Keep it up

[manya0033]

Hey @blessy05 , thanks for putting this together! I can see the GraphSAGE approach is ambitious and there's good thought behind the graph construction. A few things to address before I can approve:

The notebook is using talabat_enhanced_orders.csv from a local path, but the file isn't in the repo so reviewers can't run it. Since it's an external dataset (not from MOP), could you upload it to the DEPENDENCIES folder and update the path in the notebook accordingly, as per the PR checklist?
The notebook doesn't follow the use case template. There's no scenario, user story, introduction, dataset description, learning outcomes, or conclusion- just code cells with inline comments. Could you restructure it using the standard template? Our use cases are meant to read as step-by-step tutorials, so markdown explanations between sections are really important.

Small thing- the file should probably be renamed to follow the use case naming convention (check with the Use Case Naming Tool).

[blessy05]

Hi @manya0033 , thanks for the feedback.

I’ve updated the PR by restructuring the notebook using the required use case format, adding markdown explanations, updating the dataset path, and adding the external dataset under DEPENDENCIES.

Please review again when you get a chance. Thanks!

[molliefernandez-mentor]

Blessy, some comments from me:

• Please use the template provided in the GitHub to restructure your use case
• You need to be calling your data using an API not a CSV file.
• You need headings at each stage of the project explaining what you are doing and why, this is meant to e a tutorial explaining how to use the data.