Information systems need to be more flexible and to allow users to find content related to their context and interests. Metadata harvesting and metadata enrichments could represent a way to help users to find content and events according to their interests. However, metadata are underused and represents an interoperability challenge. This paper presents a new framework, called SMESE, and the implementation of its prototypes that consists of its semantic metadata model, a mapping ontology model and a user interest affinity model. This proposed framework makes these models interoperable with existing metadata models. 
SMESE also proposes a decision support process supporting the activation and deactivation of software features related to metadata. To consider context variability into account in modeling context-aware properties, SMESE makes use of an autonomous process that exploits context information to adapt software behavior using an enhanced metadata framework. When the user chooses preferences in terms of system behavior, the semantic weight of each feature is computed. This weight quantifies the importance of the feature for the user according to their interests.
This paper also proposed a semantic metadata analysis ecosystem to support data harvesting according to a metadata model and a mapping ontology model. Data harvesting is coupled with internal and external enrichments. The initial SMESE prototype represents more than 400 millions of relationships (triplets). To conclude, this paper also presents the design and implementation of different prototypes of SMESE applied to digital ecosystems.