Tools and process for management of network infrastructure typically assume that network administrators are technically literate, with a willingness to devote significant time to ensuring that devices are correctly configured to behave as desired. Whilst this is largely the case for service provider and enterprise networks, it is typically not the case for home area network users. The growing complexity and heterogeneity of home area networks mean that management and configuration tasks are becoming increasingly complex. In this paper we outline an approach for the semantic uplift of monitoring data from a home area network router and the selection of appropriate policies drive the (re-)configuration of the router to provide desired behaviour in response to the monitored events. We outline the use of an ontology-based semantic model to conceptualize monitoring data and select the policies to apply. The major contributions of this paper are the algorithms developed for semantic uplift and policy selection. The approach is explained and evaluated via two example scenarios that have been realized on a home area network test-bed.