July 2018
Proceedings of the 32nd International BCS Human Computer Interaction Conference (HCI)
Human Computer Interaction Conference
4 - 6 July 2018
Creativity, Creativity Support Tools, Mutual Engagement, Data Mining, Evaluation, Interaction Corpus
This paper presents a case study of using conventional Data Mining techniques to identify clusters of creative contributions made by users in a collaborative music making system. Results of the clustering suggest that when people mutually engage with each other they tend to converge on similar creative contributions, whereas smaller clusters may indicate higher quality contributions. We also show how we used Data Mining to discriminate between kinds of creative contributions including: complex melodic structures, simple melodic structures, musical motifs, and rhythmical contributions. Our use of Data Mining does not require direct interaction with users and so it may be useful in real-world study contexts.
This work is licensed under a Creative Commons Attribution 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/