Categorising the online communities of stack exchange using quantitative user behaviour features
Date
2016-03-02Author
Aumayr, Erik
Metadata
Show full item recordUsage
This item's downloads: 577 (view details)
Recommended Citation
Aumayr, Erik. (2016). Categorising the online communities of stack exchange using quantitative user behaviour features (pp. 18)
Published Version
Abstract
Maintaining online communities is vital in order to increase and retain their economic
and social value. Before applying any performance altering strategies, it is important to
determine the different types of communities, as they might be affected differently. In
the literature, we find qualitative categories such as transactional and interest-based.
However, these qualitative classification approaches do not guarantee to reflect the underlying
user behaviour. Yet it is crucial to study the user behaviour, e.g. how many
users join per day, in order to understand which communities perform well and which
ones require intervention by a community manager. In this work, we present a bottomup
community clustering approach that relies on quantitatively measurable user behaviour
features. We examine 29 online communities of the Stack Exchange platform,
and describe the extracted features that capture the user behaviour. Based on these features
we then categorise the communities. By analysing the clusters, we find that they
correspond to a certain degree to intuitive topical themes.