Header

UZH-Logo

Maintenance Infos

In search of new product ideas: Identifying ideas in online communities by machine learning and text mining


Christensen, Kasper; Nørskov, Sladjana; Frederiksen, Lars; Scholderer, Joachim (2017). In search of new product ideas: Identifying ideas in online communities by machine learning and text mining. Creativity and Innovation Management, 26(1):17-30.

Abstract

Online communities are attractive sources of ideas relevant for new product development and innovation. However, making sense of the ‘big data’ in these communities is a complex analytical task. A systematic way of dealing with these data is needed to exploit their potential for boosting companies' innovation performance. We propose a method for analysing online community data with a special focus on identifying ideas. We employ a research design where two human raters classified 3,000 texts extracted from an online community, according to whether the text contained an idea. Among the 3,000, 137 idea texts and 2,666 non-idea texts were identified. The human raters could not agree on the remaining 197 texts. These texts were omitted from the analysis. The remaining 2,803 texts were processed by using text mining techniques and used to train a classification model. We describe how to tune the model and which text mining steps to perform. We conclude that machine learning and text mining can be useful for detecting ideas in online communities. The method can help researchers and firms identify ideas hidden in large amounts of texts. Also, it is interesting in its own right that machine learning can be used to detect ideas.

Abstract

Online communities are attractive sources of ideas relevant for new product development and innovation. However, making sense of the ‘big data’ in these communities is a complex analytical task. A systematic way of dealing with these data is needed to exploit their potential for boosting companies' innovation performance. We propose a method for analysing online community data with a special focus on identifying ideas. We employ a research design where two human raters classified 3,000 texts extracted from an online community, according to whether the text contained an idea. Among the 3,000, 137 idea texts and 2,666 non-idea texts were identified. The human raters could not agree on the remaining 197 texts. These texts were omitted from the analysis. The remaining 2,803 texts were processed by using text mining techniques and used to train a classification model. We describe how to tune the model and which text mining steps to perform. We conclude that machine learning and text mining can be useful for detecting ideas in online communities. The method can help researchers and firms identify ideas hidden in large amounts of texts. Also, it is interesting in its own right that machine learning can be used to detect ideas.

Statistics

Citations

Altmetrics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:03 Faculty of Economics > Department of Informatics
Dewey Decimal Classification:000 Computer science, knowledge & systems
Language:English
Date:9 December 2017
Deposited On:22 Feb 2017 16:32
Last Modified:23 Feb 2017 04:08
Publisher:Wiley-Blackwell Publishing, Inc.
ISSN:0963-1690
Publisher DOI:https://doi.org/10.1111/caim.12202
Other Identification Number:merlin-id:14596

Download

Full text not available from this repository.
View at publisher

TrendTerms

TrendTerms displays relevant terms of the abstract of this publication and related documents on a map. The terms and their relations were extracted from ZORA using word statistics. Their timelines are taken from ZORA as well. The bubble size of a term is proportional to the number of documents where the term occurs. Red, orange, yellow and green colors are used for terms that occur in the current document; red indicates high interlinkedness of a term with other terms, orange, yellow and green decreasing interlinkedness. Blue is used for terms that have a relation with the terms in this document, but occur in other documents.
You can navigate and zoom the map. Mouse-hovering a term displays its timeline, clicking it yields the associated documents.

Author Collaborations