Linking Motif Sequences to Tale Types by Machine Learning

Abstract units of narrative content called motifs constitute sequences, also known as tale types.However whereas the dependency of tale types on the constituent motifs is clear, the strength oftheir bond has not been measured this far. Based on the observation that differences betweensuch motif sequences are reminiscent of nucleotide and chromosome mutations in genetics, i.e.,constitute “narrative DNA”, we used sequence mining methods from bioinformatics to learn moreabout the nature of tale types as a corpus. 94% of the Aarne-Thompson-Uther catalogue (2249tale types in 7050 variants) was listed as individual motif strings based on the Thompson MotifIndex, and scanned for similar subsequences. Next, using machine learning algorithms, we builtand evaluated a classifier which predicts the tale type of a new motif sequence. Our findingsindicate that, due to the size of the available samples, the classification model was best able topredict magic tales, novelles and jokes