A novel DL-based algorithm integrating medical knowledge graph and doctor modeling for Q&A pair matching in OHP.

•Automatic Q&A matching through OHP can reduce the time cost of doctor-patient interaction.•Provide the Medical KG Attention with Doctor Modeling Neural Network (MKGA-DM-NN) to solve the medical Q&A text matching task.•Use Node2vec technology and attention mechanism to embed the knowledge map into the model to guide Q&A pair matching.•Model doctors for the answers selection task, where doctor representations are learned by the key answer texts of the doctor.