Utilizing Ensemble Approach for Predictive Customer Clustering Analysis with Unsupervised Cluster Labeling

Research output: Contribution to conferencePaperpeer-review

Abstract

Customer clustering is an unsupervised machinelearning approach that groups diverse customers based on shared
characteristics. This research explores the use of machine learning, particularly the k-means model and predictive algorithms,
to improve the precision of customer cluster analysis in the
retail sector. An Ensemble approach is proposed to gain deeper
insights into customer behavior and predict future behavior
within the same clusters. The study evaluates multiple machine
learning techniques, splitting a time series dataset into two
frames, and employs K-fold cross-validation for model performance enhancement. Key evaluation metrics include Accuracy,
Precision, Recall, and F1-score. The findings show that the
Extreme Gradient Boosting Classifier performs best in Dataset
One, while Random Forest outperforms in Dataset Two. The
project aims to combine these top-performing ensemble classifiers
using the Voting Classifier to classify customers. The proposed
model achieves high precision scores for both datasets, making
it a promising approach for customer classification.
Original languageEnglish
Publication statusPublished - 2023

Fingerprint

Dive into the research topics of 'Utilizing Ensemble Approach for Predictive Customer Clustering Analysis with Unsupervised Cluster Labeling'. Together they form a unique fingerprint.

Cite this