Towards a Window-based Diverse Entity Summarisation Engine in Publish/Subscribe Systems

The rise of Smart Homes, Smart Cities and Internet of Things results in the creation of a wide range of entity-based data streams and users interested in the real-time analysis of these streams. These smart environments possess characteristics, like dynamism, continuity, heterogeneity and high volume of data and users. A suitable data dissemination paradigm is needed that can overcome these challenges and at the same time, provide expressive notifications to users, but not at the expense of usability or resources. Publish/Subscribe systems can efficiently realise some of these requirements; however, they need additional support when applied in smart environments to overcome assumptions related to usability and redundancy-awareness. Therefore, the key question of the paper is: Can we define an entity-centric Publish/Subscribe system that provides expressive user notifications along with high usability and limited resource usage? In this work, we explore this question and propose a Publish/Subscribe system with windowing, data fusion, and top-k diverse ranking that can result in the creation of expressive entity summaries using limited resources. Our results show that sending a top-k fused diverse summary as a notification is better than sending all the separate notifications or the fused ones without top-k filtering. Specifically, top-k fused diverse summarisation results in 50% to 80% reduction of forwarded messages and redundancy-awareness with an F-score ranging from 0.35 to 0.73 depending on the k. Nevertheless, these results are achieved at the expense of a slightly higher latency; therefore, there is some trade-off between latency, the number of forwarded messages, and expressiveness.