A Comparison of Mobile Rule Engines for Reasoning on Semantic Web Based Health Data

Semantic Web technology is used extensively in the health domain, due to its ability to specify expressive, domain-specific data, as well as its capacity to facilitate data integration between heterogeneous, health-related sources. In the health domain, mobile devices are an essential part of patient self-management approaches, where local clinical decision support is applied to ensure that patients receive timely clinical findings. Currently, increases in mobile device capabilities have enabled the deployment of Semantic Web technologies on mobile platforms, enabling the consumption of rich, semantically described health data. To make this semantic health data available to local decision support as well, Semantic Web reasoning should be deployed on mobile platforms. However, there is currently a lack of software solutions and performance analysis of mobile, Semantic Web reasoning engines. This paper presents and compares the mobile benchmarks of 4 reasoning engines, applied on a dataset and rule-set for patients with Atrial Fibrillation (AF). In particular, these benchmarks investigate the scalability of the mobile reasoning processes, and study reasoning performance for different process flows in decision support. For the purpose of these benchmarks, we extended a number of existing rule engines and RDF stores with Semantic Web reasoning capabilities.