Share This Author
Knowledge graph prediction of unknown adverse drug reactions and validation in electronic health records
High-confidence predictions were validated in electronic records significantly more frequently than random models, and outperformed standard methods (logistic regression, decision trees and support vector machines).
Falcon-AO: Results for OAEI 2007
A brief overview of Falcon-AO (version 0.7): a practical ontology matching system with acceptable to very good performance, a flexible architecture, and a number of unique features.
Exploiting Linked Data and Knowledge Graphs in Large Organisations
This book identifies the gaps between the requirements of enterprise knowledge consumption and standard data consuming technologies by analysing real-world use cases, and proposes the enterprise knowledge graph to fill such gaps.
SemEHR: A general-purpose semantic search system to surface semantic data from clinical notes for tailored care, trial recruitment, and clinical research*
SemEHR provides a more comprehensive view of a patient, bringing in more and unexpected insight compared to study-oriented bespoke information extraction systems.
ADEPt, a semantically-enriched pipeline for extracting adverse drug events from free-text electronic health records
A rule-based ADE detection and classification pipeline built and tested on a large Psychiatric corpus comprising 264k patients using the de-identified EHRs of four UK-based psychiatric hospitals, and presents an improvement to the state of the art context-discerning algorithm, ConText.
Improving RNN with Attention and Embedding for Adverse Drug Reactions
The impact of initializing the training of a neural network natural language processing algorithm with pre-defined clinical word embeddings to improve feature extraction and relationship classification between entities is studied.
Graph Pattern Based RDF Data Compression
- Jeff Z. Pan, José Manuél Gómez-Pérez, Y. Ren, Honghan Wu, Haofen Wang, Man Zhu
- Computer ScienceJIST
- 9 November 2014
Evaluation on real world datasets shows that the proposed graph pattern based technologies can substantially reduce the size of RDF documents by complementing the abilities of existing approaches, and the evaluation results on rule mining operations show the potentials of the proposed serialisation format in supporting efficient data access.
Big smog meets web science: smog disaster analysis based on social media and device data on the web
- Jiaoyan Chen, Huajun Chen, Guozhou Zheng, Jeff Z. Pan, Honghan Wu, Ningyu Zhang
- Computer ScienceWWW
- 7 April 2014
Social web data and device web data are integrated to build standard health hazard rating reference and train smog-health models for health hazard prediction and applied to online and location-sensitive smog disaster monitoring, which can better guide people's behaviour and government's strategy design for disaster mitigation.
Searching Semantic Web Objects Based on Class Hierarchies
The Fal- cons system, a keyword-based search engine for the Semantic Web, and a class recommendation technique is proposed to enable hierarchically navigating classes to support the class-based query restriction.
Design and Application of Rule Based Access Control Policies
This paper introduces how to express access control policies based on OWL and SWRL, which includes a definition of OWL ontology to describe the terms and a declaration of SWRL rules to explicit the relationship between properties.