Overview / Usage

Problem :

The problem is to predict the type of customer visiting a hotel based on various attributes related to the customer and the hotel. The classification of the customer type could be a contract, transient, group, etc. The accurate prediction of the customer type can help hotel managers and marketers to customize their services and marketing strategies to meet the specific needs of their customers.

Solution

The solution is to create a machine learning model that can predict the customer type based on the available data. The data preprocessing steps involve data cleaning, feature engineering, and normalization. The model selection process includes choosing appropriate algorithms such as SVM, Logistic Regression, MultinomialNB, KNeighborsClassifier, GradientBoostingClassifier, GaussianNB, DecisionTreeClassifier. The evaluation metrics for the models could be accuracy, precision, recall, and F1 score. The feature importance analysis could reveal which attributes have the most significant impact on the prediction. The trained model can be used to predict the customer type of new customers. This project provides a reference for similar classification problems in the hospitality industry and beyond.

Conclusion of the analysis

As we worked with the different types of models that have trained for predicting the solution of the problem, DecisionTreeClassifier model predict more accurate and can able to classify the customer.

Methodology / Approach

We have used several number of models to train and get the better accuracy for the particular model for the problem statement.

Models used

• svm
• Logistic Regression
• MultinomialNB
• KNeighborsClassifier
• GradientBoostingClassifier
• GaussianNB
• DecisionTreeClassifier

Technologies Used

1. OneAPI
2. Jupyter Lab (Intel One API 2023) Kernal
3. OneDAL

Documents and Presentations

Repository

https://github.com/RSVignesh06/Intel-oneAPI---Customer-type-prediction.git