Related Experiment Video
Updated: Sep 10, 2025
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
4.1K
Identifying the Intents Behind Website Visits by Employing Unsupervised Machine Learning Models
Judah Soobramoney1, Retius Chifurira1, Temesgen Zewotir1
1School of Mathematics, Statistics and Computer Science, University of KwaZulu-Natal, Durban, South Africa.
Annals of Data Science
|August 26, 2025
Summary
This study used unsupervised machine learning to analyze website visitor data, identifying five key user intents. Hierarchical clustering proved most effective for understanding online visit motivations.
Area of Science:
- Data Science
- Machine Learning
- Consumer Behavior Analysis
Background:
- Increasing digitization necessitates corporate understanding of website usage.
- Analyzing complex, high-volume website data to understand consumer behavior is challenging.
Purpose of the Study:
- To apply unsupervised machine learning models to identify visitor intentions on a corporate website.
- To extract actionable insights from complex website visit data.
Main Methods:
- Utilized Google Analytics data from a corporate informative website.
- Employed k-means, hierarchical, and DBSCAN unsupervised machine learning models.
- Evaluated model performance based on cluster homogeneity and size.
Main Results:
- All three models identified five distinct visitor intents: "accidentals", "drop-offs", "engrossed", "get-in-touch", and "seekers".
- Hierarchical clustering demonstrated superior performance in balancing cluster homogeneity and size.
Conclusions:
- Unsupervised machine learning effectively identifies user intents from website data.
- Hierarchical clustering offers a robust method for analyzing visitor motivations and optimizing corporate web strategies.
