Using Mixture Models to Characterize the Process Durations of Daily Living

Eman  Shaikh; Bryan Scotney; Sally I McClean; Zeeshan Tariq; Nazeeruddin Mohammad

Using Mixture Models to Characterize the Process Durations of Daily Living

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

28 Downloads (Pure)

Abstract

In recent years, process mining (PM) has found widespread
use across the healthcare, education, logistics, and finance domain. Smart
homes employ PM to examine human behavior, health conditions, and
enhance daily living. Existing research uses PM to study human behavior. However, it failed to provide a comprehensive approach that studied/compared the different mixture models (MM) to determine the best
model that closely characterizes human behavior. As a result, this paper
uses the gamma, Weibull and Gaussian MMs to represents the process
durations of daily living to facilitate an accurate representation of human
behavior. The Expectation-Maximization (EM) algorithm was employed
where the Kolmogorov-Smirnov (KS), Kullback-Leibler (KL) divergence,
and Cramer-von Mises (CvM) tests were chosen to determine the best
MM. The proposed approach was applied over the Kasteren, UCI and
4TU dataset.

Original language	English
Title of host publication	24th UK Workshop in Computational Intelligence (UKCI 2025)
Publication status	Accepted/In press - 1 Jul 2025
Event	24th UK Workshop in Computational Intelligence - Edinburgh, United Kingdom Duration: 3 Sept 2025 → 5 Sept 2025

Conference

Conference	24th UK Workshop in Computational Intelligence
Country/Territory	United Kingdom
City	Edinburgh
Period	3/09/25 → 5/09/25

Funding

This research is supported by the VCRS (Vice-Chancellor’s Research Studentships), funded by Ulster University.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Access to Document

Shaikh (3)Author Accepted Manuscript, 2.88 MB

Cite this

@inproceedings{61a7d149b4d64a48bb82f8845365a601,

title = "Using Mixture Models to Characterize the Process Durations of Daily Living",

abstract = "In recent years, process mining (PM) has found widespread use across the healthcare, education, logistics, and finance domain. Smart homes employ PM to examine human behavior, health conditions, and enhance daily living. Existing research uses PM to study human behavior. However, it failed to provide a comprehensive approach that studied/compared the different mixture models (MM) to determine the best model that closely characterizes human behavior. As a result, this paper uses the gamma, Weibull and Gaussian MMs to represents the process durations of daily living to facilitate an accurate representation of human behavior. The Expectation-Maximization (EM) algorithm was employed where the Kolmogorov-Smirnov (KS), Kullback-Leibler (KL) divergence, and Cramer-von Mises (CvM) tests were chosen to determine the best MM. The proposed approach was applied over the Kasteren, UCI and 4TU dataset.",

author = "Eman Shaikh and Bryan Scotney and McClean, \{Sally I\} and Zeeshan Tariq and Nazeeruddin Mohammad",

year = "2025",

month = jul,

day = "1",

language = "English",

booktitle = "24th UK Workshop in Computational Intelligence (UKCI 2025)",

note = "24th UK Workshop in Computational Intelligence ; Conference date: 03-09-2025 Through 05-09-2025",

}

TY - GEN

T1 - Using Mixture Models to Characterize the Process Durations of Daily Living

AU - Shaikh, Eman

AU - Scotney, Bryan

AU - McClean, Sally I

AU - Tariq, Zeeshan

AU - Mohammad, Nazeeruddin

PY - 2025/7/1

Y1 - 2025/7/1

N2 - In recent years, process mining (PM) has found widespread use across the healthcare, education, logistics, and finance domain. Smart homes employ PM to examine human behavior, health conditions, and enhance daily living. Existing research uses PM to study human behavior. However, it failed to provide a comprehensive approach that studied/compared the different mixture models (MM) to determine the best model that closely characterizes human behavior. As a result, this paper uses the gamma, Weibull and Gaussian MMs to represents the process durations of daily living to facilitate an accurate representation of human behavior. The Expectation-Maximization (EM) algorithm was employed where the Kolmogorov-Smirnov (KS), Kullback-Leibler (KL) divergence, and Cramer-von Mises (CvM) tests were chosen to determine the best MM. The proposed approach was applied over the Kasteren, UCI and 4TU dataset.

AB - In recent years, process mining (PM) has found widespread use across the healthcare, education, logistics, and finance domain. Smart homes employ PM to examine human behavior, health conditions, and enhance daily living. Existing research uses PM to study human behavior. However, it failed to provide a comprehensive approach that studied/compared the different mixture models (MM) to determine the best model that closely characterizes human behavior. As a result, this paper uses the gamma, Weibull and Gaussian MMs to represents the process durations of daily living to facilitate an accurate representation of human behavior. The Expectation-Maximization (EM) algorithm was employed where the Kolmogorov-Smirnov (KS), Kullback-Leibler (KL) divergence, and Cramer-von Mises (CvM) tests were chosen to determine the best MM. The proposed approach was applied over the Kasteren, UCI and 4TU dataset.

M3 - Conference contribution

BT - 24th UK Workshop in Computational Intelligence (UKCI 2025)

T2 - 24th UK Workshop in Computational Intelligence

Y2 - 3 September 2025 through 5 September 2025

ER -

Using Mixture Models to Characterize the Process Durations of Daily Living

Abstract

Conference

Funding

UN SDGs

Access to Document

Fingerprint

Cite this