This project is focused on analysing and predicting data from hotel satisfaction surveys and creating a prediction model that can forecast guest's satisfaction levels accurately based on pertinent parameters. The project is built upon the extraction, aggregation, and pre-processing of Europe Hotel Satisfaction Score data from AWS Cloud.
-
I gathered, consolidated, and pre-processed data on Hotel Satisfaction Scores across Europe from both MySQL and AWS Cloud sources. This facilitated an in-depth analysis of key features influencing satisfaction and the prediction of overall satisfaction scores.
-
Employing Python with Scikit-learn, PySpark, Tableau, and SPSS Modeler, I conducted a comprehensive examination and visualization of the dataset.
-
Rigorous data pre-processing, cleaning, and feature selection techniques were applied to enhance data quality.
-
Multiple supervised learning models were developed to discern patterns in relevant features and forecast overall guest satisfaction levels.
-
The study incorporated data mining algorithms such as Logistic Regression, Decision Trees, and Random Forests.
-
The models achieved an impressive accuracy rate exceeding 90%, underscoring their robust predictive capabilities.
-
I put forth actionable recommendations for hotel companies and the tourism industry based on predicted guest satisfaction levels. These insights can be utilized by hotels to optimize services, adjust environments, boost tourist occupancy rates, drive local economic development, and contribute to the sustainable growth of the global tourism economy.
All source code for this project is provided in a Jupyter Notebook file. To get started, clone the repository and open the notebook in Jupyter.