Classifying Diagrams and Their Parts using Graph Neural Networks: A Comparison of Crowd-Sourced and Expert Annotations
This work addresses diagram classification for educational resources, but it is incremental as it compares existing annotation methods without introducing new techniques.
The study compared crowd-sourced and expert annotations for classifying diagrams and their parts using graph neural networks, finding that layout features effectively predict element identity, while expert annotations yield better diagram type representations.
This article compares two multimodal resources that consist of diagrams which describe topics in elementary school natural sciences. Both resources contain the same diagrams and represent their structure using graphs, but differ in terms of their annotation schema and how the annotations have been created - depending on the resource in question - either by crowd-sourced workers or trained experts. This article reports on two experiments that evaluate how effectively crowd-sourced and expert-annotated graphs can represent the multimodal structure of diagrams for representation learning using various graph neural networks. The results show that the identity of diagram elements can be learned from their layout features, while the expert annotations provide better representations of diagram types.