Data-Science-Assignment-eCommerce-Transactions-Dataset

This repository contains the deliverables for the Data Science Intern assignment, focusing on analyzing eCommerce transaction data. The assignment is divided into three key tasks: Exploratory Data Analysis (EDA), Lookalike Model development, and Customer Segmentation using clustering techniques. Each task aims to derive insights and provide actionable recommendations for business improvement.

---

Overview of the Assignment

The eCommerce Transactions dataset consists of three files:

• Customers.csv: Contains customer profiles, including ID, name, region, and signup date.
• Products.csv: Lists product details such as ID, name, category, and price.
• Transactions.csv: Contains transaction data, including transaction ID, customer ID, product ID, date, quantity, and total value.

Objectives:

1. Perform EDA to uncover insights about customers, products, and sales trends.
2. Develop a Lookalike Model to identify similar customers based on profile and transaction data.
3. Use clustering techniques for customer segmentation and derive actionable insights.

---

Task Descriptions

1. Exploratory Data Analysis (EDA)

• Objective: Analyze the dataset to uncover key business insights, such as customer distribution, sales trends, and top-performing products.
• Deliverables:
  • Anvita_Magarde_EDA.ipynb: Jupyter Notebook containing the EDA process.
  • Anvita_Magarde_EDA.pdf: PDF report summarizing insights (maximum 500 words).

2. Lookalike Model

• Objective: Build a model to recommend the top 3 most similar customers for a given customer based on their profile and transaction history.
• Deliverables:
  • Anvita_Magarde_Lookalike.ipynb: Notebook explaining the model development.
  • Anvita_Magarde_Lookalike.csv: CSV file containing lookalike recommendations and similarity scores.

3. Customer Segmentation (Clustering)

• Objective: Use clustering algorithms to group customers into segments based on profile and transaction data. Evaluate clusters using metrics like the Davies-Bouldin Index.
• Deliverables:
  • Anvita_Magarde_Clustering.ipynb: Notebook containing clustering analysis and visualization.
  • Anvita_Magarde_Clustering.pdf: PDF report summarizing clustering results.

---

Instructions to Run Each Notebook

1. Clone the repository:

   git clone https://github.com/yourusername/eCommerce-Transactions-Analysis.git
   cd eCommerce-Transactions-Analysis

2. Install dependencies:

   pip install -r requirements.txt

3. Open and execute the Jupyter Notebooks:

   jupyter notebook

4. Run each notebook in the following order:

   • Anvita_Magarde_EDA.ipynb
   • Anvita_Magarde_Lookalike.ipynb
   • Anvita_Magarde_Clustering.ipynb

5. Review the generated outputs in the Outputs folder.

---

Prerequisites and Dependencies

• Python 3.8+
• Libraries:
  • pandas
  • numpy
  • matplotlib
  • seaborn
  • scikit-learn
  • jupyter