Data analytics of call log data to identify caller behaviour patterns from a mental health and wellbeing helpline

Siobhan O'Neill, RR Bond, A Grigorash, Colette Ramsey, C Armour, Maurice Mulvenna

Research output: Contribution to journalArticle

5 Citations (Scopus)
33 Downloads (Pure)

Abstract

This work presents an analysis of 3.5 million calls made to a mental health and well-being helpline, seeking to answer the question, what different groups of callers can be characterised by specific usage patterns? Calls were extracted from a telephony informatics system. Each call was logged with a date, time, duration and a unique identifier allowing for repeat caller analysis. We utilized data mining techniques to reveal new insights into help-seeking behaviours. Analysis was carried out using unsupervised machine learning (K-means clustering) to discover the types of callers, and Fourier transform was used to ascertain periodicity in calls. Callers can be clustered into five or six caller groups that offer a meaningful interpretation. Cluster groups are stable and re-emerge regardless of which year is considered. The volume of calls exhibits strong repetitive intra-day and intra-week patterns. Intra-month repetitions are absent. This work provides new data-driven findings to model the type and behaviour of callers seeking mental health support. It offers insights for computer-mediated and telephony-based helpline management.

Original languageEnglish
Pages (from-to)1722-1738
Number of pages17
JournalHealth Informatics Journal
Volume25
Issue number4
Early online date17 Sep 2018
DOIs
Publication statusPublished - 1 Dec 2019

Fingerprint

Mental Health
Informatics
Data Mining
Periodicity
Fourier Analysis
Cluster Analysis
Unsupervised Machine Learning
Help-Seeking Behavior

Keywords

  • clustering methods
  • Fourier series
  • Fourier transform
  • frequency estimation
  • healthcare service usage
  • help-seeking behaviour
  • machine learning
  • mental health
  • mental health and well-being helpline
  • psychology
  • suicide
  • telephony analysis
  • well-being

Cite this

@article{807fd95291db410480a83649ef8133a4,
title = "Data analytics of call log data to identify caller behaviour patterns from a mental health and wellbeing helpline",
abstract = "This work presents an analysis of 3.5 million calls made to a mental health and well-being helpline, seeking to answer the question, what different groups of callers can be characterised by specific usage patterns? Calls were extracted from a telephony informatics system. Each call was logged with a date, time, duration and a unique identifier allowing for repeat caller analysis. We utilized data mining techniques to reveal new insights into help-seeking behaviours. Analysis was carried out using unsupervised machine learning (K-means clustering) to discover the types of callers, and Fourier transform was used to ascertain periodicity in calls. Callers can be clustered into five or six caller groups that offer a meaningful interpretation. Cluster groups are stable and re-emerge regardless of which year is considered. The volume of calls exhibits strong repetitive intra-day and intra-week patterns. Intra-month repetitions are absent. This work provides new data-driven findings to model the type and behaviour of callers seeking mental health support. It offers insights for computer-mediated and telephony-based helpline management.",
keywords = "clustering methods, Fourier series, Fourier transform, frequency estimation, healthcare service usage, help-seeking behaviour, machine learning, mental health, mental health and well-being helpline, psychology, suicide, telephony analysis, well-being",
author = "Siobhan O'Neill and RR Bond and A Grigorash and Colette Ramsey and C Armour and Maurice Mulvenna",
year = "2019",
month = "12",
day = "1",
doi = "10.1177/1460458218792668",
language = "English",
volume = "25",
pages = "1722--1738",
journal = "Health Informatics Journal",
issn = "1460-4582",
number = "4",

}

Data analytics of call log data to identify caller behaviour patterns from a mental health and wellbeing helpline. / O'Neill, Siobhan; Bond, RR; Grigorash, A; Ramsey, Colette; Armour, C; Mulvenna, Maurice.

In: Health Informatics Journal, Vol. 25, No. 4, 01.12.2019, p. 1722-1738.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data analytics of call log data to identify caller behaviour patterns from a mental health and wellbeing helpline

AU - O'Neill, Siobhan

AU - Bond, RR

AU - Grigorash, A

AU - Ramsey, Colette

AU - Armour, C

AU - Mulvenna, Maurice

PY - 2019/12/1

Y1 - 2019/12/1

N2 - This work presents an analysis of 3.5 million calls made to a mental health and well-being helpline, seeking to answer the question, what different groups of callers can be characterised by specific usage patterns? Calls were extracted from a telephony informatics system. Each call was logged with a date, time, duration and a unique identifier allowing for repeat caller analysis. We utilized data mining techniques to reveal new insights into help-seeking behaviours. Analysis was carried out using unsupervised machine learning (K-means clustering) to discover the types of callers, and Fourier transform was used to ascertain periodicity in calls. Callers can be clustered into five or six caller groups that offer a meaningful interpretation. Cluster groups are stable and re-emerge regardless of which year is considered. The volume of calls exhibits strong repetitive intra-day and intra-week patterns. Intra-month repetitions are absent. This work provides new data-driven findings to model the type and behaviour of callers seeking mental health support. It offers insights for computer-mediated and telephony-based helpline management.

AB - This work presents an analysis of 3.5 million calls made to a mental health and well-being helpline, seeking to answer the question, what different groups of callers can be characterised by specific usage patterns? Calls were extracted from a telephony informatics system. Each call was logged with a date, time, duration and a unique identifier allowing for repeat caller analysis. We utilized data mining techniques to reveal new insights into help-seeking behaviours. Analysis was carried out using unsupervised machine learning (K-means clustering) to discover the types of callers, and Fourier transform was used to ascertain periodicity in calls. Callers can be clustered into five or six caller groups that offer a meaningful interpretation. Cluster groups are stable and re-emerge regardless of which year is considered. The volume of calls exhibits strong repetitive intra-day and intra-week patterns. Intra-month repetitions are absent. This work provides new data-driven findings to model the type and behaviour of callers seeking mental health support. It offers insights for computer-mediated and telephony-based helpline management.

KW - clustering methods

KW - Fourier series

KW - Fourier transform

KW - frequency estimation

KW - healthcare service usage

KW - help-seeking behaviour

KW - machine learning

KW - mental health

KW - mental health and well-being helpline

KW - psychology

KW - suicide

KW - telephony analysis

KW - well-being

UR - http://www.scopus.com/inward/record.url?scp=85058698578&partnerID=8YFLogxK

UR - https://pure.ulster.ac.uk/en/publications/data-analytics-of-call-log-data-to-identify-caller-behaviour-patt

U2 - 10.1177/1460458218792668

DO - 10.1177/1460458218792668

M3 - Article

C2 - 30222034

VL - 25

SP - 1722

EP - 1738

JO - Health Informatics Journal

JF - Health Informatics Journal

SN - 1460-4582

IS - 4

ER -