Predicting customer satisfaction in customer support conversations in social media using affective features

Jonathan Herzig; Guy Feigenblat; Michal Shmueli-Scheuer; David Konopnicki; Anat Rafaeli

doi:10.1145/2930238.2930285

Predicting customer satisfaction in customer support conversations in social media using affective features

Jonathan Herzig, Guy Feigenblat, Michal Shmueli-Scheuer, David Konopnicki, Anat Rafaeli

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

28 Scopus citations

Abstract

Providing customer support through social media channels is gaining popularity. In such a context, predicting customer satisfaction in an early stage of a service conversation is important. Such an analysis can help personalize agent assignment to maximize customer satisfaction, and prioritize conversations. In this paper, we show that affective features such as customer's and agent's personality traits and emotion expression improve prediction of customer satisfaction when added to more typical text based features. We only utilize information extracted from the first customer conversation turn and previous customer and agent social network activity. Thus, our customer satisfaction classifier outputs its prediction in an early stage of the conversation, before any interaction has taken place between the customer and an agent. Our model was trained and tested on a Twitter conversations dataset of two customer support services, and shows an improvement of 30% in F1-score for predicting dissatisfaction.

Original language	English
Title of host publication	UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization
Publisher	Association for Computing Machinery, Inc
Pages	115-119
Number of pages	5
ISBN (Electronic)	9781450343701
DOIs	https://doi.org/10.1145/2930238.2930285
State	Published - 13 Jul 2016
Externally published	Yes
Event	24th ACM International Conference on User Modeling, Adaptation, and Personalization, UMAP 2016 - Halifax, Canada Duration: 13 Jul 2016 → 17 Jul 2016

Publication series

Name	UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization

Conference

Conference	24th ACM International Conference on User Modeling, Adaptation, and Personalization, UMAP 2016
Country/Territory	Canada
City	Halifax
Period	13/07/16 → 17/07/16

Bibliographical note

Publisher Copyright:
© 2016 ACM.

Keywords

Affective computing
Classification
Customer support

Access to Document

10.1145/2930238.2930285

Cite this

Herzig, J., Feigenblat, G., Shmueli-Scheuer, M., Konopnicki, D., & Rafaeli, A. (2016). Predicting customer satisfaction in customer support conversations in social media using affective features. In UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization (pp. 115-119). (UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization). Association for Computing Machinery, Inc. https://doi.org/10.1145/2930238.2930285

Herzig, Jonathan ; Feigenblat, Guy ; Shmueli-Scheuer, Michal et al. / Predicting customer satisfaction in customer support conversations in social media using affective features. UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization. Association for Computing Machinery, Inc, 2016. pp. 115-119 (UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization).

@inproceedings{3d5b8506169343d79ac1025b398f4554,

title = "Predicting customer satisfaction in customer support conversations in social media using affective features",

abstract = "Providing customer support through social media channels is gaining popularity. In such a context, predicting customer satisfaction in an early stage of a service conversation is important. Such an analysis can help personalize agent assignment to maximize customer satisfaction, and prioritize conversations. In this paper, we show that affective features such as customer's and agent's personality traits and emotion expression improve prediction of customer satisfaction when added to more typical text based features. We only utilize information extracted from the first customer conversation turn and previous customer and agent social network activity. Thus, our customer satisfaction classifier outputs its prediction in an early stage of the conversation, before any interaction has taken place between the customer and an agent. Our model was trained and tested on a Twitter conversations dataset of two customer support services, and shows an improvement of 30% in F1-score for predicting dissatisfaction.",

keywords = "Affective computing, Classification, Customer support",

author = "Jonathan Herzig and Guy Feigenblat and Michal Shmueli-Scheuer and David Konopnicki and Anat Rafaeli",

note = "Publisher Copyright: {\textcopyright} 2016 ACM.; 24th ACM International Conference on User Modeling, Adaptation, and Personalization, UMAP 2016 ; Conference date: 13-07-2016 Through 17-07-2016",

year = "2016",

month = jul,

day = "13",

doi = "10.1145/2930238.2930285",

language = "אנגלית",

series = "UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization",

publisher = "Association for Computing Machinery, Inc",

pages = "115--119",

booktitle = "UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization",

}

Herzig, J, Feigenblat, G, Shmueli-Scheuer, M, Konopnicki, D & Rafaeli, A 2016, Predicting customer satisfaction in customer support conversations in social media using affective features. in UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization. UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization, Association for Computing Machinery, Inc, pp. 115-119, 24th ACM International Conference on User Modeling, Adaptation, and Personalization, UMAP 2016, Halifax, Canada, 13/07/16. https://doi.org/10.1145/2930238.2930285

Predicting customer satisfaction in customer support conversations in social media using affective features. / Herzig, Jonathan; Feigenblat, Guy; Shmueli-Scheuer, Michal et al.
UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization. Association for Computing Machinery, Inc, 2016. p. 115-119 (UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization).

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

TY - GEN

T1 - Predicting customer satisfaction in customer support conversations in social media using affective features

AU - Herzig, Jonathan

AU - Feigenblat, Guy

AU - Shmueli-Scheuer, Michal

AU - Konopnicki, David

AU - Rafaeli, Anat

PY - 2016/7/13

Y1 - 2016/7/13

N2 - Providing customer support through social media channels is gaining popularity. In such a context, predicting customer satisfaction in an early stage of a service conversation is important. Such an analysis can help personalize agent assignment to maximize customer satisfaction, and prioritize conversations. In this paper, we show that affective features such as customer's and agent's personality traits and emotion expression improve prediction of customer satisfaction when added to more typical text based features. We only utilize information extracted from the first customer conversation turn and previous customer and agent social network activity. Thus, our customer satisfaction classifier outputs its prediction in an early stage of the conversation, before any interaction has taken place between the customer and an agent. Our model was trained and tested on a Twitter conversations dataset of two customer support services, and shows an improvement of 30% in F1-score for predicting dissatisfaction.

AB - Providing customer support through social media channels is gaining popularity. In such a context, predicting customer satisfaction in an early stage of a service conversation is important. Such an analysis can help personalize agent assignment to maximize customer satisfaction, and prioritize conversations. In this paper, we show that affective features such as customer's and agent's personality traits and emotion expression improve prediction of customer satisfaction when added to more typical text based features. We only utilize information extracted from the first customer conversation turn and previous customer and agent social network activity. Thus, our customer satisfaction classifier outputs its prediction in an early stage of the conversation, before any interaction has taken place between the customer and an agent. Our model was trained and tested on a Twitter conversations dataset of two customer support services, and shows an improvement of 30% in F1-score for predicting dissatisfaction.

KW - Affective computing

KW - Classification

KW - Customer support

UR - http://www.scopus.com/inward/record.url?scp=84984904475&partnerID=8YFLogxK

U2 - 10.1145/2930238.2930285

DO - 10.1145/2930238.2930285

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:84984904475

T3 - UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization

SP - 115

EP - 119

BT - UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization

PB - Association for Computing Machinery, Inc

T2 - 24th ACM International Conference on User Modeling, Adaptation, and Personalization, UMAP 2016

Y2 - 13 July 2016 through 17 July 2016

ER -

Herzig J, Feigenblat G, Shmueli-Scheuer M, Konopnicki D, Rafaeli A. Predicting customer satisfaction in customer support conversations in social media using affective features. In UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization. Association for Computing Machinery, Inc. 2016. p. 115-119. (UMAP 2016 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization). doi: 10.1145/2930238.2930285

Predicting customer satisfaction in customer support conversations in social media using affective features

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this