Cloud-Hosted Notebook Data Manipulation

This project demonstrates data manipulation and exploratory data analysis (EDA) on a sample dataset using a cloud-hosted Jupyter Notebook on Google Colab. The notebook integrates with GitHub for version control and sharing.

---

Project Overview

This project performs EDA on a dataset containing customer purchasing behavior using Python libraries like pandas, Matplotlib, and Seaborn. It also highlights the seamless integration between Google Colab and GitHub for cloud-based data analysis and collaboration.

Collab Jupyter Notebook:

https://colab.research.google.com/github/nogibjj/Mobasserul_Haque_MiniProject9/blob/main/collab_main.ipynb

---

File Structure

├── .devcontainer/
│   ├── devcontainer.json
│   └── Dockerfile
├── .github/
│   └── workflows/
│       └── cicd.yml
├── myLib/
│   ├── __pycache__/
│   ├── __init__.py
│   ├── lib.py
├── bar_plot_average_purchase_amt_by_regions.png
├── collab_main.ipynb
├── Correlation_matrix_columns.png
├── Customer Purchasing Behaviors.csv
├── demo.mp4
├── Histogram_column_distributions.png
├── Loyalty_score_by_region_boxplot.png
├── main.ipynb
├── main.py
├── Makefile
├── README.md
├── requirements.txt
├── scatter_plot_hue_by_region.png
├── scatter_plot_trend_line.png
├── summary_report.md
├── test_lib.py
├── test_main.py
├── test_report.md                        

Setup Instructions

1. Access Google Colab

• Go to Google Colab.
• Sign in with your Google account.
• Open the notebook directly by clicking the "Open in Colab" badge above or follow these steps:
  • Create a new notebook by selecting File > New Notebook.
  • Alternatively, open an existing notebook from your GitHub repository.

2. Clone the Repository in Colab

To access all project files in Colab, clone the repository by running the following in a notebook cell:

!git clone https://github.com/nogibjj/Mobasserul_Haque_MiniProject9.git

Navigate to the Project Directory

%cd Mobasserul_Haque_MiniProject9

3. Install Dependencies

Install the required Python libraries using the requirements.txt file:

!pip install -r requirements.txt

Run the Notebook

Open the collab_main.ipynb file in Colab and run all the cells to execute the data manipulation and visualization tasks.

---

Features

• Summary Statistics: Summary statistics of customer purchasing data.
• Data Visualizations: Including histograms, scatter plots, box plots, and correlation heatmaps.
• Google Colab Integration: Run Jupyter notebooks in the cloud.

Source

The dataset used in this project is Customer Purchasing Behaviors, which can be found on Kaggle:

About Dataset

Dataset Columns

• customer_id: Unique ID of the customer.
• age: The age of the customer.
• annual_income: The customer's annual income (in USD).
• purchase_amount: The total amount of purchases made by the customer (in USD).
• purchase_frequency: Frequency of customer purchases (number of times per year).
• region: The region where the customer lives (North, South, East, West).
• loyalty_score: Customer's loyalty score (a value between 0-100).

Summary Statistics

The summary_statistics function displays key metrics like the median, range, and variance for each numerical column.

	count	mean	std	min	25%	50%	75%	max	median	range	variance
user_id	238.0	119.500000	68.848868	1.0	60.25	119.5	178.750	238.0	119.5	237.0	4.740167e+03
age	238.0	38.676471	9.351118	22.0	31.00	39.0	46.750	55.0	39.0	33.0	8.744341e+01
annual_income	238.0	57407.563025	11403.875717	30000.0	50000.00	59000.0	66750.000	75000.0	59000.0	45000.0	1.300484e+08
purchase_amount	238.0	425.630252	140.052062	150.0	320.00	440.0	527.500	640.0	440.0	490.0	1.961458e+04
loyalty_score	238.0	6.794118	1.899047	3.0	5.50	7.0	8.275	9.5	7.0	6.5	3.606379e+00
purchase_frequency	238.0	19.798319	4.562884	10.0	17.00	20.0	23.000	28.0	20.0	18.0	2.081991e+01

Data Visualization

Below are sample visualizations produced by the project:

• Distribution of Columns:

• Loyalty Score by Region:

• Scatter Plot with Hue (Region):

• Scatter Plot with Trend Line:

• Correlation Matrix:

• Bar Plot of Average Purchase Amount by Region:

Continuous Integration (CI)

The repository includes a GitHub Actions CI pipeline for automated testing, linting, and build checks.

To view the status, navigate to the Actions tab on GitHub.

---

Collaborate and Share

• Save your notebook changes to GitHub by selecting File > Save a copy in GitHub from Colab.
• Push updates directly to your GitHub repository.