Predicting Caller Type From a Mental Health and Well-Being Helpline: Analysis of Call Log Data

Alexander Grigorash, Siobhan O'Neill, Raymond Bond, Colette Ramsey, Cherie Armour, Maurice Mulvenna

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Background: This paper presents an analysis of call data records pertaining to a telephone helpline in Ireland among individuals seeking mental health and well-being support and among those who are in a suicidal crisis.Objective: The objective of our study was to examine whether rule sets generated from decision tree classification, trained using features derived from callers’ several initial calls, could be used to predict what caller type they would become.Methods: Machine learning techniques were applied to the call log data, and five distinct patterns of caller behaviors were revealed, each impacting the helpline capacity in different ways.Results: The primary findings of this study indicate that a significant model (P<.001) for predicting caller type from call log data obtained from the first 8 calls is possible. This indicates an association between callers’ behavior exhibited during initial calls and their behavior over the lifetime of using the service.Conclusions: These data-driven findings contribute to advanced workload forecasting for operational management of the telephone-based helpline and inform the literature on helpline caller behavior in general.
LanguageEnglish
Article numbere47
Number of pages14
JournalJMIR Mental Health
Volume5
Issue number2
DOIs
Publication statusPublished - 11 Jun 2018

Fingerprint

Mental Health
Telephone
Decision Trees
Workload
Ireland

Keywords

  • data mining
  • machine learning
  • clustering
  • classification
  • mental health
  • Suicide

Cite this

@article{1dc0e337fc6d4df39520f6385fc75f9a,
title = "Predicting Caller Type From a Mental Health and Well-Being Helpline: Analysis of Call Log Data",
abstract = "Background: This paper presents an analysis of call data records pertaining to a telephone helpline in Ireland among individuals seeking mental health and well-being support and among those who are in a suicidal crisis.Objective: The objective of our study was to examine whether rule sets generated from decision tree classification, trained using features derived from callers’ several initial calls, could be used to predict what caller type they would become.Methods: Machine learning techniques were applied to the call log data, and five distinct patterns of caller behaviors were revealed, each impacting the helpline capacity in different ways.Results: The primary findings of this study indicate that a significant model (P<.001) for predicting caller type from call log data obtained from the first 8 calls is possible. This indicates an association between callers’ behavior exhibited during initial calls and their behavior over the lifetime of using the service.Conclusions: These data-driven findings contribute to advanced workload forecasting for operational management of the telephone-based helpline and inform the literature on helpline caller behavior in general.",
keywords = "data mining, machine learning, clustering, classification, mental health, Suicide",
author = "Alexander Grigorash and Siobhan O'Neill and Raymond Bond and Colette Ramsey and Cherie Armour and Maurice Mulvenna",
year = "2018",
month = "6",
day = "11",
doi = "10.2196/mental.9946",
language = "English",
volume = "5",
journal = "JMIR Mental Health",
issn = "2368-7959",
number = "2",

}

Predicting Caller Type From a Mental Health and Well-Being Helpline: Analysis of Call Log Data. / Grigorash, Alexander; O'Neill, Siobhan; Bond, Raymond; Ramsey, Colette; Armour, Cherie; Mulvenna, Maurice.

In: JMIR Mental Health, Vol. 5, No. 2, e47, 11.06.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Predicting Caller Type From a Mental Health and Well-Being Helpline: Analysis of Call Log Data

AU - Grigorash, Alexander

AU - O'Neill, Siobhan

AU - Bond, Raymond

AU - Ramsey, Colette

AU - Armour, Cherie

AU - Mulvenna, Maurice

PY - 2018/6/11

Y1 - 2018/6/11

N2 - Background: This paper presents an analysis of call data records pertaining to a telephone helpline in Ireland among individuals seeking mental health and well-being support and among those who are in a suicidal crisis.Objective: The objective of our study was to examine whether rule sets generated from decision tree classification, trained using features derived from callers’ several initial calls, could be used to predict what caller type they would become.Methods: Machine learning techniques were applied to the call log data, and five distinct patterns of caller behaviors were revealed, each impacting the helpline capacity in different ways.Results: The primary findings of this study indicate that a significant model (P<.001) for predicting caller type from call log data obtained from the first 8 calls is possible. This indicates an association between callers’ behavior exhibited during initial calls and their behavior over the lifetime of using the service.Conclusions: These data-driven findings contribute to advanced workload forecasting for operational management of the telephone-based helpline and inform the literature on helpline caller behavior in general.

AB - Background: This paper presents an analysis of call data records pertaining to a telephone helpline in Ireland among individuals seeking mental health and well-being support and among those who are in a suicidal crisis.Objective: The objective of our study was to examine whether rule sets generated from decision tree classification, trained using features derived from callers’ several initial calls, could be used to predict what caller type they would become.Methods: Machine learning techniques were applied to the call log data, and five distinct patterns of caller behaviors were revealed, each impacting the helpline capacity in different ways.Results: The primary findings of this study indicate that a significant model (P<.001) for predicting caller type from call log data obtained from the first 8 calls is possible. This indicates an association between callers’ behavior exhibited during initial calls and their behavior over the lifetime of using the service.Conclusions: These data-driven findings contribute to advanced workload forecasting for operational management of the telephone-based helpline and inform the literature on helpline caller behavior in general.

KW - data mining

KW - machine learning

KW - clustering

KW - classification

KW - mental health

KW - Suicide

U2 - 10.2196/mental.9946

DO - 10.2196/mental.9946

M3 - Article

VL - 5

JO - JMIR Mental Health

T2 - JMIR Mental Health

JF - JMIR Mental Health

SN - 2368-7959

IS - 2

M1 - e47

ER -