Building Models of Installations to Recommend Applications in IoT Software Ecosystems

Internet of Things devices are driven by applications. To stimulate innovation on their platforms, manufacturers of embedded products are creating software ecosystems that enable third-parties develop applications for their devices. Since such applications are developed externally, their seamless integration into the platforms is a challenge. In this paper, we propose to use semantic reasoning to select and configure applications in software ecosystems. Our prototype builds a model of the installation and the user goals through several steps. It processes application descriptions created by their developers and matches them to the installation model. Based on the matched descriptions, it presents options to deploy applications in the environment to the user sorted by their relevance. The proposed approach enables resolving complex requirements of applications before their installation in the platform. Our preliminary evaluation shows that the prototype supports the integration of the identified applications.