Interaction Network Analysis Using Semantic Similarity Based on Translation Embeddings

Abstract

Biomedical knowledge graphs such as STITCH, SIDER, and Drugbank provide the basis for the discovery of associations between biomedical entities, e.g., interactions between drugs and targets. Link prediction is a paramount task and represents a building block for supporting knowledge discovery. Although several approaches have been proposed for effectively predicting links, the role of semantics has not been studied in depth. In this work, we tackle the problem of discovering interactions between drugs and targets, and propose SimTransE, a machine learning-based approach that solves this problem effectively. SimTransE relies on translating embeddings to model drug-target interactions and values of similarity across them. Grounded on the vectorial representation of drug-target interactions, SimTransE is able to discover novel drug-target interactions. We empirically study SimTransE using state-ofthe-art benchmarks and approaches. Experimental results suggest that SimTransE is competitive with the state of the art, representing, thus, an effective alternative for knowledge discovery in the biomedical domain.

Research on fault simulation and influence range deduction technology for backbone network communication system based on TransE improved algorithm

“…The energy function E used by DKRL consists of four sub energy functions added together. After defining such a complex energy function, the structural and semantic vectors of entities can interact and promote each other during the training process [8]. The specific expression form of energy function E is:…”