Sense making is a process where basic data usually collected from different, heterogeneous sources is set into perspective and yields new insights or a deeper understanding. In particular, sense making is generally a means to look at data in the light of a certain application domain or application-specific question. That includes the integration of data from different, possibly noisy sources, the selection of relevant data and the exclusion of irrelevant data, making different pieces of information plausible with respect to prior hypotheses, and empirically testing whether conjectures derived from initial models over samples of data sources (usually of limited size and varying quality) hold for real world data. The techniques used for sense making include knowledge representation and meta-data enhancement, Big Data analytics, uncertainty and ambiguity management and resolution, as well as a variety of reasoning techniques.

Modern data-driven intelligent applications rely on components making sense of large sets of heterogenous, diverse and possibly noisy data from different sources. In order to build those applications, we need rigorous foundations for sense making, means for semantic integration of data, extraction of narratives, assessment of data quality etc. Intelligent applications including components for sense making need engineering principles and methods considering the specific challenges of those data-driven sense making algorithms. In particular, test and verification techniques are required that ensure the quality of system results influenced by sense making. A particularly interesting application domain for sense making are large sets of geo-spatial data for analytic and predictive applications.

The WAKERS2 workshop is intended to bring together researchers in all fields of sense making, from foundations, to engineering, analysis and application.