HEALTHFINLAND is a national semantic publishing system for providing Finnish citizens with reliable, up-to-date information about health. The system consists of three parts: 1) a centralized service of health ontologies with tools, 2) a semantic content creation channel based on several distributed health organizations, and 3) an intelligent semantic portal aggegating and presenting the contents from intuitive and health promoting end-user perspectives. The system demontrates how semantic web techniques can be applied to solving problems of distributed content creation, discovery, linking, aggregation, and reuse in health information on a national level, from end-users’, content publishers’, and machine processing viewpoints. The HEALTHFINLAND prototype is operational on the web, and a production version of the portal, created by the National Health Institute, will be released in February 2009. 1 Problems of Mediating Health Information Health information is one of the most frequently searched material on the web [1]. However, a citizen searching for health information on the web faces many challenges [2]. Health information is often published at organization-centric websites, requiring prior knowledge of the organizations involved. After finding a piece of interesting information, it is often tedious and difficult to find related relevant web resources. Outdated and broken links are common. Satisfying an end-user’s information need often requires aggregation of content from several information providers. Additionally, the quality and trustworthiness of information varies. In many cases it is difficult know whether a content is based on scientific results or layman opinions and rumors, or whether it is motivated by commercial interests. Research organizations that focus on scientific issues cannot necessarily communicate the results effectively to ordinarily citizens. From the viewpoint of the health organizations, creating health information to citizens is also problematic in many ways [2]. Several organizations create overlapping content, which wastes time and money and is confusing to the end user. Content in websites is usually minimally annotated for the purpose of presenting it on a particular site and for the particular purpose of the publisher. This makes it difficult and expensive for other organizations to re-use content across portals even if the portal owners were willing to do this. The problems of maintaining links up-to-date is very costly and tedious, especially when dealing with links to external sites to which the maintainer and the content management system has no control. If metadata is produced, finding the right keywords and other metadata descriptions for web pages and documents is difficult and time consuming for information producers. The vocabularies used, such as MeSH1, UMLS2 or SNOMED CT3, are very large and require expertise to use. Furthermore, there are several quality issues involved when publishing health information: 1) Quality of the content creation process (e.g. regular reviews and updates of published material) 2) Quality of the content itself (e.g., errors in the medical subject matter or understandability for the target audience). 3) Quality of additional information on pages (e.g., it is advisable to show the date of publication on each page). 4) Quality of the metadata. For example, one indexer may use only few general keywords while another prefers a longer detailed list, which leads to problems of unbalanced and low quality metadata. HEALTHFINLAND addresses these problems both from the publishers and citizens viewpoints. Furthermore, the solutions are provided for machines to use through semantic widgets. 2 The HEALTHFINLAND Solution An overview of the HEALTHFINLAND system is given in Figure 1. The system consists of content sources such as health websites, content aggregation infrastructure, ontology services and a citizens’ portal. The novelty of the system is based on three major ideas: First, HEALTHFINLAND minimizes duplicate redundant work and costs in creating health content on a national level by collaboration. A goal of the HEALTHFINLAND collaborative production network is to ensure that information about a health topic is produced only once and by the organization that knows most about it. By using semantic technologies the content can then be re-used in different web portals by the other organizations, not only in the organization’s own web site. This possibility is facilitated by annotating the content locally with semantic metadata based on shared ontologies, and by making the global repository available by a semantic portal and as mash-up web services. This is a generalization of the idea of “multi-channel publication” of XML, where a single syntactic structure can be rendered in different ways, but on the semantic metadata level and using RDF: semantic content is re-used through multi-application publication. The second key idea behind HEALTHFINLAND is to try to minimize the maintenance costs of portals by letting the computer take care of semantic link maintenance and aggregation of content from the different publishers. This possibility is also based on shared semantic metadata and ontologies. New content relevant to a topic may be 1 http://www.nlm.nih.gov/mesh/ 2 http://umlsinfo.nlm.nih.gov 3 http://www.snomed.org/snomedct/ HealthFinland Portal Diabetes Assoc
[1]
Véronique Malaisé,et al.
A Method to Convert Thesauri to SKOS
,
2006,
ESWC.
[2]
Eero Hyvönen,et al.
Building a National Semantic Web Ontology and Ontology Service Infrastructure -The FinnONTO Approach
,
2008,
ESWC.
[3]
Eero Hyvönen,et al.
HealthFinland - Finnish Health Information on the Semantic Web
,
2007,
ISWC/ASWC.
[4]
Andy Powell.
Expressing Dublin Core in HTML/XHTML meta and link elements
,
2003
.
[5]
J. B. Brooke,et al.
SUS: A 'Quick and Dirty' Usability Scale
,
1996
.
[6]
Eero Hyvönen,et al.
Distributed Semantic Content Creation and Publication for Cultural Heritage Legacy Systems
,
2008
.
[7]
Eero Hyvönen,et al.
User-Centric Faceted Search for Semantic Portals
,
2007,
ESWC.
[8]
Eero Hyvönen,et al.
Application of Ontology Techniques to View-Based Semantic Search and Browsing
,
2004,
ESWS.
[9]
Eero Hyvönen,et al.
Publishing and Using Ontologies as Mash-Up Services
,
2008
.
[10]
Eero Hyvönen,et al.
Efficient Content Creation on the Semantic Web Using Metadata Schemas with Domain Ontology Services (System Description)
,
2007,
ESWC.