Data quality in internet of things: A state-of-the-art survey

In the Internet of Things (IoT), data gathered from a global-scale deployment of smart-things, are the base for making intelligent decisions and providing services. If data are of poor quality, decisions are likely to be unsound. Data quality (DQ) is crucial to gain user engagement and acceptance of the IoT paradigm and services. This paper aims at enhancing DQ in IoT by providing an overview of its state-of-the-art. Data properties and their new lifecycle in IoT are surveyed. The concept of DQ is defined and a set of generic and domain-specific DQ dimensions, fit for use in assessing IoT's DQ, are selected. IoT-related factors endangering the DQ and their impact on various DQ dimensions and on the overall DQ are exhaustively analyzed. DQ problems manifestations are discussed and their symptoms identified. Data outliers, as a major DQ problem manifestation, their underlying knowledge and their impact in the context of IoT and its applications are studied. Techniques for enhancing DQ are presented with a special focus on data cleaning techniques which are reviewed and compared using an extended taxonomy to outline their characteristics and their fitness for use for IoT. Finally, open challenges and possible future research directions are discussed.