Using information-seeking argument mining to improve service
- If service providers can identify reasons users are in favor of or against a service, they have insightful information that can help them understand user behavior and what they need to do to change such behavior. This article argues that the novel text-mining technique referred to as information-seeking argument mining (IS-AM) can identify these reasons. The empirical study applies IS-AM to news articles and reviews about electric scooter-sharing systems (i.e., a service enabling the short-term rentals of electric motorized scooters). Its results point to IS-AM as a promising technique to improve service; the data enable the authors to identify 40 reasons to use or not use electric scooter-sharing systems, as well as their importance to users. Furthermore, the results show that news articles are better data sources than reviews because they are longer and contain more arguments and, thus, reasons.
Author: | Bernd SkieraORCiDGND, Shunyao YanORCiDGND, Johannes DaxenbergerORCiDGND, Marcus DomboisORCiDGND, Iryna GurevychORCiDGND |
---|---|
URN: | urn:nbn:de:hebis:30:3-862801 |
DOI: | https://doi.org/10.1177/10946705221110845 |
ISSN: | 1552-7379 |
Parent Title (English): | Journal of service research |
Publisher: | Sage Periodicals Press |
Place of publication: | London |
Document Type: | Article |
Language: | English |
Date of Publication (online): | 2022/06/29 |
Date of first Publication: | 2022/06/29 |
Publishing Institution: | Universitätsbibliothek Johann Christian Senckenberg |
Release Date: | 2024/07/21 |
Tag: | argument mining; service design; service improvement; service innovation; textual analysis |
Volume: | 25 |
Issue: | 4 |
Page Number: | 12 |
First Page: | 537 |
Last Page: | 548 |
HeBIS-PPN: | 520875648 |
Institutes: | Wirtschaftswissenschaften / Wirtschaftswissenschaften |
Angeschlossene und kooperierende Institutionen / E-Finance Lab e.V. | |
Dewey Decimal Classification: | 3 Sozialwissenschaften / 33 Wirtschaft / 330 Wirtschaft |
Sammlungen: | Universitätspublikationen |
Licence (German): | ![]() |