Integrating Guided Clustering in Visual Analytics to Support Domain Expert Reasoning Processes

Abstract

Recent research shows promise in combining Information Visualization (IV) and Machine Learning (ML) to assist data analysis performed by domain experts. However, this approach presents non-trivial challenges, in particular when the goal is to incorporate knowledge provided by the domain expert in underlying ML algorithms. To address these challenges, we present an analytical process and a visual analytics tool that uses visual queries to capture examples from the domain experts' existing reasoning process which will guide the subsequent clustering. Our work is motivated by a collaboration with personnel at the Danish Business Authority, who are interested in two types of insights: (1) On which data dimensions is a selected subset of companies different from the remaining companies? (2) Which other companies lie within the same multi-dimensional subspace? The poster will illustrate a real analysis scenario, where the presented analytic process allows auditors to use their knowledge of identified "suspicious" companies to kick-start the analysis for others.