Joint Extraction of Uyghur Medicine Knowledge with Edge Computing
Fan Lu, Quan Qi, Huaibin Qin
TL;DR
The paper addresses privacy and latency challenges in extracting Uyghur traditional medical knowledge by proposing CoEx-Bert, a joint, parameter-sharing extraction model deployed on edge devices. By coupling two sub-models for entity and relation extraction with shared hidden representations and a unified loss, CoEx-Bert reduces error propagation and leverages contextual interdependence to improve accuracy. Empirical results on a Uyghur medical dataset show superior precision, recall, and F1 (0.9065, 0.9245, 0.9154 respectively) over strong baselines, with favorable edge-device performance (65 ms latency, 100 MB bandwidth). The approach also demonstrates generalization to related languages (e.g., Mongolian texts) and highlights practical benefits for real-time, privacy-preserving medical knowledge extraction, with future work aimed at enhancing named entity recognition via integration with high-performing NER models.
Abstract
Medical knowledge extraction methods based on edge computing deploy deep learning models on edge devices to achieve localized entity and relation extraction. This approach avoids transferring substantial sensitive data to cloud data centers, effectively safeguarding the privacy of healthcare services. However, existing relation extraction methods mainly employ a sequential pipeline approach, which classifies relations between determined entities after entity recognition. This mode faces challenges such as error propagation between tasks, insufficient consideration of dependencies between the two subtasks, and the neglect of interrelations between different relations within a sentence. To address these challenges, a joint extraction model with parameter sharing in edge computing is proposed, named CoEx-Bert. This model leverages shared parameterization between two models to jointly extract entities and relations. Specifically, CoEx-Bert employs two models, each separately sharing hidden layer parameters, and combines these two loss functions for joint backpropagation to optimize the model parameters. Additionally, it effectively resolves the issue of entity overlapping when extracting knowledge from unstructured Uyghur medical texts by considering contextual relations. Finally, this model is deployed on edge devices for real-time extraction and inference of Uyghur medical knowledge. Experimental results demonstrate that CoEx-Bert outperforms existing state-of-the-art methods, achieving accuracy, recall, and F1 scores of 90.65\%, 92.45\%, and 91.54\%, respectively, in the Uyghur traditional medical literature dataset. These improvements represent a 6.45\% increase in accuracy, a 9.45\% increase in recall, and a 7.95\% increase in F1 score compared to the baseline.