Mapping WordNet onto human brain connectome in emotion processing and semantic similarity recognition

Jan Kocon, Marek Maziarz

Abstract: Highlights • Novel representation of heterogeneous combined graph knowledge bases. • Interlinked multilingual wordnets and human brain connectome graph. • Mapping procedure from synset categories to behavioural functions of brain regions. • New method for calculating embeddings for elements of graph knowledge base. • Improved method of propagating information in a graph with deep neural network. • Improved propagation of emotions in a wordnet expanded with brain-based connections. Abstract In this article we extend a WordNet structure with relations linking synsets to Desikan’s brain regions. Based on lexicographer files and WordNet Domains the mapping goes from synset semantic categories to behavioural and cognitive functions and then directly to brain lobes. A human brain connectome (HBC) adjacency matrix was utilised to capture transition probabilities between brain regions. We evaluated the new structure in several tasks related to semantic similarity and emotion processing using brain-expanded Princeton WordNet (207k LUs) and Polish WordNet (285k LUs, 30k annotated with valence, arousal and 8 basic emotions). A novel HBC vector representation turned out to be significantly better than proposed baselines.