Problem Statement

"Trips & Travel.Com" company wants to enable and establish a viable business model to expand the customer base. One of the ways to expand the customer base is to introduce a new offering of packages. Currently, there are 5 types of packages the company is offering - Basic, Standard, Deluxe, Super Deluxe, King. Looking at the data of the last year, we observed that 18% of the customers purchased the packages. However, the marketing cost was quite high because customers were contacted at random without looking at the available information. The company is now planning to launch a new product i.e. Wellness Tourism Package. Wellness Tourism is defined as Travel that allows the traveler to maintain, enhance or kick-start a healthy lifestyle, and support or increase one's sense of well-being. However, this time company wants to harness the available data of existing and potential customers to make the marketing expenditure more efficient.

About Dataset:

What's inside is more than just rows and columns.

• Most important features that have an impact on Product taken: Designation, Passport, Tier City, Martial status, occupation
• Customers with Designation as Executive should be the target customers for the company .Customers who have passport and are from tier 3 city and are single or unmarried, have large business such customers have higher chances of taking new package.
• Customers monthly income in range of 15000- 25000, and age range 15-30, prefer 5 star properties also have higher chances of taking new package based on EDA.

We need to analyze the customers' data and information to provide recommendations to the Policy Maker and Marketing Team and also build a model to predict the potential customer who is going to purchase the newly introduced travel package.

Tasks to Solve :

• To predict which customer is more likely to purchase the newly introduced travel package
• Which variables are most significant.
• Which segment of customers should be targeted more.

Dataset used:

"Travel.csv"

📊 Model Performance Comparison

This section compares the performance of Random Forest, SVM, and XGBoost models before and after hyperparameter tuning.
The evaluation is based on standard classification metrics: Accuracy, Precision, Recall, F1-Score, and ROC-AUC.

---

🔹 Before Hyperparameter Tuning

Model	Accuracy	Precision	Recall	F1-Score	ROC-AUC
Random Forest	0.9294	0.9832	0.6359	0.7723	0.9638
SVM	0.8160	0.5079	0.6957	0.5872	0.8467
XGBoost	0.9121	0.8712	0.6250	0.7278	0.9360

---

🔹 After Hyperparameter Tuning

Model	Accuracy	Precision	Recall	F1-Score	ROC-AUC
Random Forest (Tuned)	0.9294	0.9832	0.6359	0.7723	0.9638
SVM (Tuned)	0.9264	0.7979	0.8152	0.8065	0.9537
XGBoost (Tuned)	0.9356	0.9007	0.7391	0.8119	0.9606

---

📈 Improvement Summary

• Random Forest

  • Already well-optimized before tuning
  • Maintained extremely high Precision and best ROC-AUC
  • Conservative model with high-confidence predictions

• SVM

  • Significant improvement after tuning
  • Large gains in Recall and F1-Score
  • Became competitive with ensemble models

• XGBoost

  • Achieved the best overall balance
  • Highest Accuracy and F1-Score after tuning
  • Strong and consistent performance across all metrics

---

🧠 Key Observations

• Hyperparameter tuning substantially improved SVM and XGBoost performance.
• Random Forest remained the most precise and best at ranking (highest ROC-AUC).
• XGBoost (Tuned) offered the best trade-off between Precision and Recall.
• Model choice depends on business objectives:
  • High precision / low false positives → Random Forest
  • Balanced performance → XGBoost
  • High recall / fewer missed positives → SVM

---

✅ Final Model Selection

• Selected Model: XGBoost (Tuned)
• Reason: Best overall balance with highest Accuracy and F1-Score while maintaining strong ROC-AUC.

---

📝 Notes

• All models were trained using a unified preprocessing pipeline.
• Evaluation was performed on a held-out test set.
• ROC-AUC was used as the primary optimization metric during hyperparameter tuning.