HOLMeS: eHealth in the Big Data and Deep Learning Era

摘要: Now, data collection and analysis are becoming more important in a variety of application domains, as long novel technologies advance. At the same time, we experiencing growing need for human–machine interaction with expert systems, pushing research toward new knowledge representation models paradigms. In particular, last few years, eHealth—which usually indicates all healthcare practices supported by electronic elaboration remote communications—calls availability smart environment big computational resources able to offer advanced analytics human–computer The aim this paper is introduce HOLMeS (health online medical suggestions) system: A particular platform aiming at supporting several eHealth applications. As its main novelty/functionality, exploits machine learning algorithm, deployed on cluster-computing environment, order provide suggestions via both chat-bot web-app modules, especially prevention aims. chat-bot, opportunely trained leveraging deep approach, helps overcome limitations cold between users software, exhibiting human-like behavior. obtained results demonstrate effectiveness algorithms, showing an area under ROC (receiver operating characteristic) curve (AUC) 74.65% when some first-level features used assess occurrence different chronic diseases within specific pathways. When disease-specific added, shows AUC 86.78%, achieving greater clinical decisions.