Loading...
Paper Number
1347
Paper Type
Completed
Description
The early identification of new and auspicious ideas leads to competitive advantages for companies. Thereby, topic modeling can serve as an effective analytical approach for the automated investigation of trends from unstructured social media data. However, existing trend analysis tools do not meet the requirements regarding (a) Product Development, (b) Customer Behavior Analysis, and (c) Market-/Brand-Monitoring as reflected within extant literature. Thus, based on the requirements for each of these common marketing-related use cases, we derived design principles following design science research and instantiated the artifact “MANTRA” (MArketiNg TRend Analysis). We demonstrated MANTRA on a real-world data set (~1.03 million Yelp reviews) and hereby could confirm remarkable trends of vegan and global cuisine. In particular, the importance of meeting all specific requirements of the respective use cases and especially flexibly incorporating several external parameters into the trend analysis is exemplified.
Recommended Citation
Wörner, Janik; Konadl, Daniel; Schmid, Isabel Maria; and Leist, Susanne, "MANTRA: A Topic Modeling-Based Tool to Support Automated Trend Analysis on Unstructured Social Media Data" (2023). ICIS 2023 Proceedings. 15.
https://aisel.aisnet.org/icis2023/dab_sc/dab_sc/15
MANTRA: A Topic Modeling-Based Tool to Support Automated Trend Analysis on Unstructured Social Media Data
The early identification of new and auspicious ideas leads to competitive advantages for companies. Thereby, topic modeling can serve as an effective analytical approach for the automated investigation of trends from unstructured social media data. However, existing trend analysis tools do not meet the requirements regarding (a) Product Development, (b) Customer Behavior Analysis, and (c) Market-/Brand-Monitoring as reflected within extant literature. Thus, based on the requirements for each of these common marketing-related use cases, we derived design principles following design science research and instantiated the artifact “MANTRA” (MArketiNg TRend Analysis). We demonstrated MANTRA on a real-world data set (~1.03 million Yelp reviews) and hereby could confirm remarkable trends of vegan and global cuisine. In particular, the importance of meeting all specific requirements of the respective use cases and especially flexibly incorporating several external parameters into the trend analysis is exemplified.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
13-DataAnalytics