-
Our team <3 (from left to right: Xander, James, Rachel, Harvey)
Check out our presentation here:
https://pitch.com/public/43beb03c-aa89-4983-81ab-6c08b79769d9
Introduction
Welcome to our project on hotel cancellation prediction and business recommendations. Our team has developed a machine learning classification model that predicts hotel cancellations and offers business recommendations to help lower cancellation rates. In this project, we provide an overview of our approach, the tools we used, and the challenges we faced.
What it does
Our machine-learning model analyzes hotel booking data and predicts the probability of a booking being cancelled. Our system determines the importance of specific models, allowing us to provide business recommendations to help reduce cancellation rates.
How we built it
To build our machine learning model, we used a variety of data science tools including Python, Pandas, NumPy, Seaborn, Matplotlib, and Scikit-learn. Using the provided hotel booking data from various sources, we then preprocessed the data, performed feature engineering to extract useful features from the data, and split the data into training and testing sets.
Next, we built and fine-tuned our machine-learning model. We experimented with various algorithms such as Logistic Regression, Decision Trees, and Random Forests. We also optimized hyperparameters using cross-validation to improve our model's accuracy.
Finally, with the predictions on the likelihood of cancellations and considering the importance of each feature on our model, we provided business recommendations on how to reduce cancellations.
Challenges we ran into
As a team relatively new to data science, we faced several challenges throughout the project. One of the biggest challenges was feature engineering, especially dealing with a mix of categorical and continuous data. We also struggled with hyperparameter tuning and selecting the most appropriate algorithm for our problem.
Accomplishments that we're proud of
While balancing this project with our busy schedules, we're extremely proud of being able to successfully build a machine-learning model that predicted hotel cancellations with such a high degree of accuracy in a field our team was unfamiliar with.
What we learned
Throughout this project, we learned many valuable skills related to data science and machine learning. We gained experience in feature engineering, hyperparameter tuning, and working with a mix of categorical and continuous data. We also learned how to experiment with different machine-learning algorithms and optimize our model's performance.
Built With
- machine-learning
- numpy
- pandas
- python
- scikit
- scikit-learn
- seaborn
Log in or sign up for Devpost to join the conversation.