State-of-the-art approaches in the field of neural-embedding models
(NEMs) enable progress in the automatic extraction and prediction of semantic
relations between important entities like active substances, diseases, and genes.
In particular, the prediction property is making them valuable for important
research-related tasks such as hypothesis generation and drug-repositioning. A
core challenge in the biomedical domain is to have interpretable semantics from
NEMs that can distinguish, for instance, between the following two situations:
a) drug x induces disease y and b) drug x treats disease y. However, NEMs
alone cannot distinguish between associations such as treats or induces. Is it
possible to develop a model to learn a latent representation from the NEMs
capable of such disambiguation? To what extent do we need domain knowledge
to succeed in the task? In this paper, we answer both questions and show
that our proposed approach not only succeeds in the disambiguation task but
also advances current growing research e orts to nd real predictions using a
sophisticated retrospective analysis. Furthermore, we investigate which type
of associations are generally better contextualized and therefore probably have
a stronger influence in our disambiguation task. In this context, we present
an approach to extract an interpretable latent semantic subspace from the
original embedding space in which therapeutic drug-disease associations are
more likely.