Semantic Querying for Data Science on the Edge

Semantic Querying for Data Science on the Edge: From Home Automation to Personal Data Store

In the future, we envision smart home solutions to not only perform home automation but also to act as personal data stores. For example, sensing a service model envisions to create a personal data marketplace where data consumers would be able to request data from a large number of data owners. In this work, We look at this problem from the data science perspective. More specifically, we demonstrate how semantic web technologies can be used to perform ‘data wrangling’ at the edge in an efficient and effective way to help data consumers to avoid acquiring unnecessary data. We developed an ontology by re-purposing several well-known ontologies such as Semantic Sensor Network (SSN) to model data gathered from a variety of different Internet of Things (IoT) products and services. We prototype the proposed approach using an open-source home automation software, OpenHab. Through a series of use case scenarios, we demonstrate how increasingly complex data wrangling tasks can be performed with semantic web technologies. We hope our work will motivate data scientists to utilise semantic data modelling and integration technique as a viable option within the data science pipeline. Our approach also helps to protect user privacy through minimised data acquisition. We also discuss several lessons learnt and challenges identified related to spatial and temporal personal data modelling at the edge.

Team

Funding