Journey Segmentation of Turkish Tobacco Users Using Sequence Clustering Techniques

Ahmet Talha Yiğit*, Tolga Kaya, Utku Doğruak

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


In this paper, using individual tobacco panel data, a novel and behavioral approach based on sequence clustering techniques is proposed to have a deeper understanding of different behavior types of Turkish tobacco users during the consecutive price markups of 6th April and 2nd May in 2019. To achieve this, main brands before markups are determined for each of the 5052 individuals. Then, having some prior assumptions, their purchase behaviors are obtained as time-stamped event sequences. Finally, while a portion of the obtained sequences which are less complex are segmented with empirical analyses, the rest of them are segmented using hierarchical clustering with optimal matching event (OME) distance. Results suggested seven main type of behavior among the tobacco users in Turkey during the markup period.

Original languageEnglish
Title of host publicationIntelligent and Fuzzy Techniques
Subtitle of host publicationSmart and Innovative Solutions - Proceedings of the INFUS 2020 Conference
EditorsCengiz Kahraman, Sezi Cevik Onar, Basar Oztaysi, Irem Ucal Sari, Selcuk Cebi, A. Cagri Tolga
Number of pages8
ISBN (Print)9783030511555
Publication statusPublished - 2021
EventInternational Conference on Intelligent and Fuzzy Systems, INFUS 2020 - Istanbul, Turkey
Duration: 21 Jul 202023 Jul 2020

Publication series

NameAdvances in Intelligent Systems and Computing
Volume1197 AISC
ISSN (Print)2194-5357
ISSN (Electronic)2194-5365


ConferenceInternational Conference on Intelligent and Fuzzy Systems, INFUS 2020

Bibliographical note

Publisher Copyright:
© 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG.


  • Hierarchical clustering
  • Individual tobacco panel
  • Sequence analysis
  • Sequence clustering
  • Unsupervised learning


Dive into the research topics of 'Journey Segmentation of Turkish Tobacco Users Using Sequence Clustering Techniques'. Together they form a unique fingerprint.

Cite this