Gift recommendation with multilabel clustering

Abstract

In the rapidly evolving e-commerce landscape, personalized gift recommendation systems play a crucial role in enhancing customer satisfaction and driving sales. This study introduces a gift recommendation system using a multilabel clustering approach with the using four algorithms, aiming to provide personalized and accurate product suggestions. The proposed system compares the performance of four algorithms: K-nearest neighbors (K-NN), decision trees, random forest, and eXtreme gradient boosting (XGBoost). Through extensive model training and hyperparameter tuning, XGBoost demonstrated superior performance with a label ranking average precision score of 95% and minimal overfitting, outperforming other algorithms in accuracy and runtime. The results highlight the effectiveness of XGBoost in managing complex data and delivering precise recommendations, making it a valuable tool for improving user experience and increasing revenue in e-commerce platforms.