University of Twente Student Theses

Login

Profiling of potential higher education website visitors based on online behaviours: A machine learning approach

Gupta, Parth (2018) Profiling of potential higher education website visitors based on online behaviours: A machine learning approach.

[img] PDF
1MB
Abstract:Purpose: The objective of this paper is to discover the behavioural profiles of website users in the domain of higher education. Design: In this research, a framework is developed which defines the process regarding the use of unsupervised machine learning algorithms in multiple stages for a variety of datasets which differs in terms of volume, ability to handle dimensionality, type (categorical/numeric) and its availability in R language. Findings: Outcomes from the application of proposed framework reveals that machine learning algorithms created the meaningful behavioural profiles as well as captured the minute differences between them. Research limitations: This research is limited by volume and veracity of the dataset used. Practical implications: This study will help a marketer to ameliorate targeting of the advertising campaigns. Also, it empowers the SMEs (Small and medium-sized enterprises) to efficiently execute the behavioural targeting under tight budget constraints or limited resources. Originality/value: To the best of the researcher's knowledge, in the domain of higher education, none of the studies used complete linkage (hierarchical clustering) in combination with K-modes. This methodology-oriented approach renders direction to create meaningful clusters for a small-scale symmetric binary dataset with low dimensionality.
Item Type:Essay (Master)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:85 business administration, organizational science
Programme:Business Administration MSc (60644)
Link to this item:http://purl.utwente.nl/essays/75902
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page