Supervised Machine Learning Project

Problem Statement

Fyntra is one of the largest online clothing companies in the USA. It sells clothing online but also provides in-store styling and clothing advice sessions. Customers visit the store, have sessions with a personal stylist, and later order their chosen clothes either through a mobile app or a website.

The company wants to decide whether to focus on improving the mobile app experience or continue investing in the website. As a drastic measure, it is also considering shutting down the website.

As a Machine Learning expert, your role is to help the company make an informed decision by analyzing sales trends, customer interactions, and revenue patterns.

Objective

The project aims to analyze customer behavior and revenue generation based on interactions across platforms (*website vs. mobile app) using *Supervised Machine Learning techniques.

---

Tasks Performed

1. Data Analysis & Correlation Studies

   • Created joint plots using Seaborn to compare:
     • Time on Website vs. Yearly Amount Spent
     • Time on App vs. Yearly Amount Spent
   • Identified which medium (app or website) has a stronger correlation with revenue.

2. Exploratory Data Analysis (EDA)

   • Studied relationships between features in the dataset.
   • Identified key influencing factors for Yearly Amount Spent.

3. Model Building

   • Created a Linear Regression Model with:
     • Length of Membership and Yearly Amount Spent as key predictors.
   • Checked if the data fits well in the linear model.

4. Model Training & Evaluation

   • Trained and tested the model using random_state=85.
   • Evaluated model accuracy using:
     • Predictions vs. Actual Data
     • Root Mean Squared Error (RMSE)

5. Business Insights

   • Based on *model coefficients, provided insights on whether the company should invest more in its *mobile app or keep focusing on the website.

Technologies Used

• Python
• Pandas
• NumPy
• Seaborn
• Matplotlib
• Scikit-Learn (Linear Regression Model)

Dataset

• The dataset contains customer purchase behavior, time spent on platforms, and yearly expenditure.
• Used for analyzing trends and building predictive models.

Results & Conclusion

• Identified whether customer spending is influenced more by the website or the mobile app.
• Provided a data-driven recommendation for Fyntra’s business strategy.

How to Run the Project

1. Clone this repository:

   git clone https://github.com/your-username/Supervised-ML-Project.git
   

2. Navigate to the project folder:

cd Supervised-ML-Project

3. Install required libraries:

pip install -r requirements.txt

4. Open and run supervised_ML.ipynb in Jupyter Notebook.

Future Scope

Extend the model to include customer demographics and marketing strategies.

Improve prediction accuracy using advanced regression models.