Abstract
In the context of patients communicating about their disease, there are several channels along which this can be done. Most of these channels do not take the patient as primary input, but provide authoritative information. The Narrator system supplies patients with information extracted from personal stories in plain text format called "narratives". These will be processed and stored using techniques from both Information Retrieval and Natural Language Processing. As such, the system will be set up as a toolbox implementing different approaches while a Service Oriented Architecture provides the framework for integration. In this paper such approaches are described together with efforts to combine them within a suitable architecture. Furthermore, some of the important implementation details are discussed. As a starting point for the system, experiments have been carried out with initial narratives, the results of which are discussed.
Get full access to this article
View all access options for this article.