Visual Human+Machine Learning

Abstract

Abstract-In this paper we describe a novel method to integrate interactive visual analysis and machine learning to support the insight generation of the user. The suggested approach combines the vast search and processing power of the computer with the superior reasoning and pattern recognition capabilities of the human user. An evolutionary search algorithm has been adapted to assist in the fuzzy logic formalization of hypotheses that aim at explaining features inside multivariate, volumetric data. Up to now, users solely rely on their knowledge and expertise when looking for explanatory theories. However, it often remains unclear whether the selected attribute ranges represent the real explanation for the feature of interest. Other selections hidden in the large number of data variables could potentially lead to similar features. Moreover, as simulation complexity grows, users are confronted with huge multidimensional data sets making it almost impossible to find meaningful hypotheses at all. We propose an interactive cycle of knowledge-based analysis and automatic hypothesis generation. Starting from initial hypotheses, created with linking and brushing, the user steers a heuristic search algorithm to look for alternative or related hypotheses. The results are analyzed in information visualization views that are linked to the volume rendering. Individual properties as well as global aggregates are visually presented to provide insight into the most relevant aspects of the generated hypotheses. This novel approach becomes computationally feasible due to a GPU implementation of the time-critical parts in the algorithm. A thorough evaluation of search times and noise sensitivity as well as a case study on data from the automotive domain substantiate the usefulness of the suggested approach.

Representative Factor Generation for the Interactive Visual Analysis of High-Dimensional Data

“…Our method, differently, enables the integration of the computational tools by interactively determining local domains where these tools are then applied on. Fuchs et al [13] integrate methods from machine learning with interactive visual analysis to assist the user in knowledge discovery. Oeltze et al [28] demonstrate how statistical methods, such as correlation analysis and principal component analysis, are used interactively to assist the derivation of new features in the analysis of multivariate data.…”

Multiverse Data-Flow Control

“…For example when cloning a cursor, a monitor is automatically added. In future work we would like to explore a machine assisted analysis process which integrates machine learning [FWG09], parameter space sampling [BM10] and interactive visual exploration for multiverse analysis.…”

A Comparison of Pareto Sets and Jacobi Sets

“…See Lampe et al [17] for an approach to transform volumes into curves in a neighborhood-preserving fashion such that the curves can be laid out and analyzed in parallel. For an evolutionary approach, Fuchs et al [12] used an evolutionary search algorithm to assist hypothesis formalization in multiple volumetric data.…”