subedinab / amazon-ml

Women's Clothing Reviews Dataset Analysis

Project Overview

This project involves ingesting, transforming, and analyzing the Women's Clothing Reviews dataset using AWS data stack services. The main steps include:

1. Ingest and Transform Dataset:

   • List the dataset files in the public S3 bucket.
   • Copy the data locally to the notebook.
   • Load and preview the data using Pandas.
   • Transform the data to create a new CSV file with selected columns.

2. Register Dataset for Querying and Visualization:

   • Register the transformed dataset into an S3-backed database table for querying.
   • Use AWS Glue and Amazon Athena to register and query the dataset.

3. Data Analysis with AWS Data Wrangler:

   • Analyze the dataset using AWS Data Wrangler.
   • Plot visuals to extract insights from the dataset.

Project Structure

The project is organized into the following sections:

1. Ingest and Transform Dataset:

   • File: 01_ingest_transform.ipynb
   • Description: This notebook includes code to list, copy, and transform the dataset.

2. Register Dataset for Querying and Visualization:

   • File: 02_register_dataset.ipynb
   • Description: This notebook demonstrates how to register the dataset in the AWS Glue Catalog.

3. Data Analysis with AWS Data Wrangler:

   • File: 03_data_analysis.ipynb
   • Description: This notebook showcases data analysis using AWS Data Wrangler.

4. Additional Files:

   • womens_clothing_ecommerce_reviews.csv: Original dataset.
   • womens_clothing_ecommerce_reviews_transformed.csv: Transformed dataset.
   • README.md: Project documentation.

Running the Notebooks

To run the notebooks, follow these steps:

1. Open 01_ingest_transform.ipynb and execute the cells to ingest and transform the dataset.
2. Open 02_register_dataset.ipynb and execute the cells to register the dataset in the AWS Glue Catalog.
3. Open 03_data_analysis.ipynb and execute the cells for data analysis.

Dependencies

Ensure you have the necessary dependencies installed, including the AWS CLI, Pandas, AWS Data Wrangler, and Boto3. You can install them using:

pip install pandas awswrangler boto3