Fig 1.
A flow diagram describing the search strategy used to identify papers for the creativity corpus.
Fig 2.
Representation of the disciplinary breakdown of the Creativity Corpus by time period.
Disciplines are as specified for the paper’s journal, by the academic database Scopus. Note that Scopus may classify a journal under more than one discipline.
Fig 3.
A flow diagram describing the search strategy used to identify papers for the non-creativity corpus.
Table 1.
The top 20 results (in descending order) of the log-likelihood ratio (LLR) calculations.
A significant LLR score at p = 0.001 is 10.83. N.B. POS = Part Of Speech: N = noun, J = adjective, V = verb, R = adverb.
Fig 4.
Word pairwise similarity data visualised as an edge-weighted graph.
Nodes correspond to words and edges are weighted by similarity scores (for any score > 0).
Fig 5.
Sample of clusters produced by the Chinese Whispers clustering step.
Fig 6.
Illustration of the process of using manual inspection for further clustering.
Fig 7.
The fourteen key components of creativity identified through an analysis of the word clusters.
Fig 8.
The ontology of creativity generated from this work’s results, in graph form.