This study examines the issues of privacy protection, data security, and query efficiency in blockchain-based electronic medical record (EMR) sharing. It proposes a secure storage and sharing scheme for EMR based on Hyperledger Fabric and the InterPlanetary File System (IPFS). To mitigate the privacy risks of data mining that could reveal patient identities, we establish an attribution channel in Hyperledger Fabric to store EMR ownership information and a data channel to store the storage location, digest, and usage records of medical data. Encrypted medical data are stored in the IPFS. To improve query efficiency in the blockchain, we integrate queryable medical data attributes into a composite key for conditional queries, avoiding complex data filtering processes. Additionally, we use a zero-knowledge proof combined with smart contracts for decentralized identity verification, eliminating reliance on third-party centralized verification services and enhancing system security. We also integrate AES and proxy re-encryption techniques to ensure data security during sharing. This scheme provides a more secure, efficient, and privacy-preserving approach for EMR systems, with significant practical implications and broad application potential.
In the field of public opinion analysis, sentiment analysis is an important basic research branch. Previous studies have successfully proved that the advanced transformer pre‐training model can be applied to this scenario in Uyghur and other low‐resource language scenarios. However, the majority of these studies are based on the traditional language anchor point and rely on the pre‐training model's cross‐lingual understanding ability. The Senti‐eXLM model proposed in this paper employs a method that allows for adaptively expanding the model's knowledge domain and dynamically adjusting the model for Uyghur language in order to improve the language's understanding and representation capability, thereby increasing the accuracy of text emotion analysis. Experiments on publicly available data sets demonstrate that when compared to the original model, the model's emotion classification accuracy is improved by 6.17%, the training convergence speed is increased by 27%, and the average reasoning time is increased by only 11%.
25 September 2025 in Expert Systems with Applications