Search

Services for semantic text annotation enable developers to extract meaning through mentioned named entities. They provide good results and their quality and usability is constantly improving. Therefore these services are becoming a frequently used component for implementations of advanced features in web sites or content management systems, e.g. recommendation systems or search engines. The goal of this thesis is implementation of semantic annotation service of english text with help of linked open data and open source software. Service should provide good enough results and, in the case of further development, represent a viable alternative to commercial services. This thesis briefly introduces domains of semantic text annotation, semantic web and linked open data. It describes software and data components of proprietary implementation of semantic text annotation service. It presents suggested text processing architecture and service implementation details. There is a simple comparison of proprietary service to two commercial services and their results are graphically shown and discussed. Further development options are presented. The thesis concludes with proprietary development experiences and a brief forecast about semantic text annotation services.