Active learning for conversational interfaces in healthcare applications

Aki Härmä; Andrey Polyakov; Ekaterina Artemova

doi:10.1007/978-3-030-12738-1_4

Active learning for conversational interfaces in healthcare applications

Aki Härmä^*
, Andrey Polyakov
, Ekaterina Artemova

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic

Abstract

In automated health services based on text and voice interfaces, there is a need to be able to understand what the user is talking about, and what is the attitude of the user towards a subject. Typical machine learning methods for text analysis require a lot of annotated data for the training. This is often a problem in addressing specific and possibly very personal health care needs. In this paper, we propose an active learning algorithm for the training of a text classifier for a conversational therapy application in the area of health behavior change. A new active learning algorithm, Query by Embedded Committee (QBEC), is proposed in the paper. The methods are particularly suitable for the text classification task in a dynamic environment and give a good performance with realistic test data.

Original language	English
Title of host publication	Artificial Intelligence in Health - 1st International Workshop, AIH 2018, Revised Selected Papers
Editors	Pau Herrero, Andrew Koster, Fernando Koch, Isabelle Bichindaritz
Publisher	Springer-Verlag London Ltd.
Pages	48-58
Number of pages	11
ISBN (Print)	9783030127374
DOIs	https://doi.org/10.1007/978-3-030-12738-1_4
Publication status	Published - 2019
Externally published	Yes
Event	1st International Workshop on Artificial Intelligence in Health - Stockholm, Sweden Duration: 13 Jul 2018 → 14 Jul 2018 Conference number: 1

Publication series

Series	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11326 LNAI
ISSN	0302-9743

Workshop

Workshop	1st International Workshop on Artificial Intelligence in Health
Abbreviated title	AIH 2018
Country/Territory	Sweden
City	Stockholm
Period	13/07/18 → 14/07/18

Access to Document

10.1007/978-3-030-12738-1_4

Cite this

Härmä, Aki ; Polyakov, Andrey ; Artemova, Ekaterina. / Active learning for conversational interfaces in healthcare applications. Artificial Intelligence in Health - 1st International Workshop, AIH 2018, Revised Selected Papers. editor / Pau Herrero ; Andrew Koster ; Fernando Koch ; Isabelle Bichindaritz. Springer-Verlag London Ltd., 2019. pp. 48-58 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11326 LNAI).

@inbook{6b1718ff273346ffbbea933e3be07059,

title = "Active learning for conversational interfaces in healthcare applications",

abstract = "In automated health services based on text and voice interfaces, there is a need to be able to understand what the user is talking about, and what is the attitude of the user towards a subject. Typical machine learning methods for text analysis require a lot of annotated data for the training. This is often a problem in addressing specific and possibly very personal health care needs. In this paper, we propose an active learning algorithm for the training of a text classifier for a conversational therapy application in the area of health behavior change. A new active learning algorithm, Query by Embedded Committee (QBEC), is proposed in the paper. The methods are particularly suitable for the text classification task in a dynamic environment and give a good performance with realistic test data.",

author = "Aki H{\"a}rm{\"a} and Andrey Polyakov and Ekaterina Artemova",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2019.; 1st International Workshop on Artificial Intelligence in Health, AIH 2018 ; Conference date: 13-07-2018 Through 14-07-2018",

year = "2019",

doi = "10.1007/978-3-030-12738-1\_4",

language = "English",

isbn = "9783030127374",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer-Verlag London Ltd.",

pages = "48--58",

editor = "Pau Herrero and Andrew Koster and Fernando Koch and Isabelle Bichindaritz",

booktitle = "Artificial Intelligence in Health - 1st International Workshop, AIH 2018, Revised Selected Papers",

}

Härmä, A, Polyakov, A & Artemova, E 2019, Active learning for conversational interfaces in healthcare applications. in P Herrero, A Koster, F Koch & I Bichindaritz (eds), Artificial Intelligence in Health - 1st International Workshop, AIH 2018, Revised Selected Papers. Springer-Verlag London Ltd., Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11326 LNAI, pp. 48-58, 1st International Workshop on Artificial Intelligence in Health, Stockholm, Sweden, 13/07/18. https://doi.org/10.1007/978-3-030-12738-1_4

Active learning for conversational interfaces in healthcare applications. / Härmä, Aki; Polyakov, Andrey; Artemova, Ekaterina.
Artificial Intelligence in Health - 1st International Workshop, AIH 2018, Revised Selected Papers. ed. / Pau Herrero; Andrew Koster; Fernando Koch; Isabelle Bichindaritz. Springer-Verlag London Ltd., 2019. p. 48-58 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11326 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic

TY - CHAP

T1 - Active learning for conversational interfaces in healthcare applications

AU - Härmä, Aki

AU - Polyakov, Andrey

AU - Artemova, Ekaterina

N1 - Conference code: 1

PY - 2019

Y1 - 2019

N2 - In automated health services based on text and voice interfaces, there is a need to be able to understand what the user is talking about, and what is the attitude of the user towards a subject. Typical machine learning methods for text analysis require a lot of annotated data for the training. This is often a problem in addressing specific and possibly very personal health care needs. In this paper, we propose an active learning algorithm for the training of a text classifier for a conversational therapy application in the area of health behavior change. A new active learning algorithm, Query by Embedded Committee (QBEC), is proposed in the paper. The methods are particularly suitable for the text classification task in a dynamic environment and give a good performance with realistic test data.

AB - In automated health services based on text and voice interfaces, there is a need to be able to understand what the user is talking about, and what is the attitude of the user towards a subject. Typical machine learning methods for text analysis require a lot of annotated data for the training. This is often a problem in addressing specific and possibly very personal health care needs. In this paper, we propose an active learning algorithm for the training of a text classifier for a conversational therapy application in the area of health behavior change. A new active learning algorithm, Query by Embedded Committee (QBEC), is proposed in the paper. The methods are particularly suitable for the text classification task in a dynamic environment and give a good performance with realistic test data.

UR - https://www.scopus.com/pages/publications/85062572141

U2 - 10.1007/978-3-030-12738-1_4

DO - 10.1007/978-3-030-12738-1_4

M3 - Chapter

AN - SCOPUS:85062572141

SN - 9783030127374

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 48

EP - 58

BT - Artificial Intelligence in Health - 1st International Workshop, AIH 2018, Revised Selected Papers

A2 - Herrero, Pau

A2 - Koster, Andrew

A2 - Koch, Fernando

A2 - Bichindaritz, Isabelle

PB - Springer-Verlag London Ltd.

T2 - 1st International Workshop on Artificial Intelligence in Health

Y2 - 13 July 2018 through 14 July 2018

ER -

Härmä A, Polyakov A, Artemova E. Active learning for conversational interfaces in healthcare applications. In Herrero P, Koster A, Koch F, Bichindaritz I, editors, Artificial Intelligence in Health - 1st International Workshop, AIH 2018, Revised Selected Papers. Springer-Verlag London Ltd. 2019. p. 48-58. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11326 LNAI). doi: 10.1007/978-3-030-12738-1_4

Active learning for conversational interfaces in healthcare applications

Abstract

Publication series

Workshop

Access to Document

Other files and links

Fingerprint

Cite this