Abstract
Raw source data can be made accessible in the form of processable information through Data-as-a-Service (DaaS) architectures. Machine learning is one possible way that allows to produce meaningful information and knowledge based on this raw source data. Thus, quality is a major concern that applies to raw data as well as to information provided by ML-generated models. Quality management is a major concern of AI Engineering – an attempt to systematically produce quality AI solutions. As the core of our solution, we define a conceptual framework that links input data quality and the machine learned data service quality, specifically inferring raw data problems as root causes from observed data service deficiency symptoms. This will allow to identify the hidden origins of quality problems that might be observed by users of DaaS offerings. We analyse the quality framework using a real-world case study from an edge cloud and Internet-of-Things-based traffic application. We identify quality assessment techniques for symptom and cause analysis.