This repository contains the code for a machine learning model that is trained on the ACME Happiness Survey dataset provided by Apziva. The goal of the model is to predict customer satisfaction based on survey responses.

Prerequisites

The following packages are required to run the code:

1. pandas
2. numpy
3. matplotlib and seaborn
4. scikit-learn

You can install these packages by running the following command: pip install -r requirements.txt

Train and predict

I found that both the KNN and the Decision Tree classifying algorithms perform equally well in predicting the target label. The code for the KNN model can be found in the train_knn.py file. The code for the Decision Tree model can be found in the train_dt.py file. To train the model, run the following command: python [file]

The model will be trained on the ACME-HappinessSurvey2020.csv dataset and the accuracy will be printed.

Similarly the code for using the trained model to make predictions on new data is contained in the predict_knn.py and predict_dt.py for the KNN and Decision Tree classifing algorithms, respectively.

Evaluation

The model uses algorithm with accuracy_score and f1_score as evaluation metric.

Conclusion

In this project, I leveraged the power of machine learning and employed the decision tree recursive algorithm to construct a robust, feature-efficient, and highly accurate ML classifier. The main goal was to predict the customers' satisfaction of the services provided. The model achieves an accuracy of 74% with both the KNN and Decision Tree models.