The State of the Art in Integrating Machine Learning into Visual Analytics

Endert, Alex; Ribarsky, William; Turkay, Çağatay; Wong, William; Nabney, Ian T.; Díaz, Ignacio; Rossi, Fabrice

doi:10.1111/cgf.13092

Cited by 176 publications

(122 citation statements)

References 162 publications

(222 reference statements)

Supporting

Mentioning

122

Contrasting

Order By: Relevance

“…Since the term Visual Analytics (VA) was coined in 2004 [WT04, TC05], there have been hundreds of software systems reported in the literature for supporting various VA workflows (e.g., [ERT∗17, LGH∗17,CLY17]). As articulated by Thomas and Kielman [TK09] and Robertson et.…”

Section: Related Workmentioning

confidence: 99%

An Ontological Framework for Supporting the Design and Evaluation of Visual Analytics Systems

Chen

Ebert

2019

Computer Graphics Forum

View full text Add to dashboard Cite

Designing, evaluating, and improving visual analytics (VA) systems is a primary area of activities in our discipline. In this paper, we present an ontological framework for recording and categorizing technical shortcomings to be addressed in a VA workflow, reasoning about the causes of such problems, identifying technical solutions, and anticipating secondary effects of the solutions. The methodology is built on the theoretical premise that designing a VA workflow is an optimization of the cost‐benefit ratio of the processes in the workflow. It makes uses three fundamental measures to group and connect “symptoms”, “causes”, “remedies”, and “side‐effects”, and guide the search for potential solutions to the problems. In terms of requirement analysis and system design, the proposed methodology can enable system designers to explore the decision space in a structured manner. In terms of evaluation, the proposed methodology is time‐efficient and complementary to various forms of empirical studies, such as user surveys, controlled experiments, observational studies, focus group discussions, and so on. In general, it reduces the amount of trial‐and‐error in the lifecycle of VA system development.

show abstract

Section: Related Workmentioning

confidence: 99%

An Ontological Framework for Supporting the Design and Evaluation of Visual Analytics Systems

Chen

Ebert

2019

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…See Figure 1. The above discussion only scratches the surface in this multidisciplinary research area, for more comprehensive selection of work, good starting points are review papers by Endert et al [1] and Sacha et al [19].…”

Section: A Few Examples Of Visualization and Machine Learning Integramentioning

confidence: 99%

“…Users interact with interactive systems in distinctive ways and often the successful execution of an analysis session depends on the accurate identification of user intent [1]. Algorithms can be trained on user activity logs to understand the user better and can offer "personalized" analytical recommendations to improve the data analysis process.…”

Section: Challenges and Opportunities For Researchmentioning

confidence: 99%

“…Knowledge generated from data drives innovation in almost all application domains, including health, transport, cyber security, manufacturing, digital services, and also scientific domains such as biology, medicine, environmental and physical sciences, the humanities and social sciences to name a few [1]. The archived data, however, is becoming increasingly complex and heterogeneous, and making sense of such data collections is becoming increasingly challenging.…”

Section: Introductionmentioning

confidence: 99%

“…Expert users have the domain knowledge to steer algorithmic power to where it is needed the most [4], and offer the capability and creativity to fine-tune computational results and turn them into data-informed decisions [5]. As a growing field, there are already several effective examples where the combination of visualisation and machine learning are being developed to offer novel solutions for data-intensive problems [1].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the Challenges and Opportunities in Visualization for Machine Learning and Knowledge Extraction: A Research Agenda

Turkay¹,

Laramee²,

Holzinger³

2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Interpretability of machine learning‐based prediction models in healthcare

Štiglic

Kocbek

Fijačko

et al. 2020

WIREs Data Min & Knowl

240

143

View full text Add to dashboard Cite

There is a need of ensuring that learning (ML) models are interpretable. Higher interpretability of the model means easier comprehension and explanation of future predictions for end-users. Further, interpretable ML models allow healthcare experts to make reasonable and data-driven decisions to provide personalized decisions that can ultimately lead to higher quality of service in healthcare. Generally, we can classify interpretability approaches in two groups where the first focuses on personalized interpretation (local interpretability) while the second summarizes prediction models on a population level (global interpretability). Alternatively, we can group interpretability methods into model-specific techniques, which are designed to interpret predictions generated by a specific model, such as a neural network, and model-agnostic approaches, which provide easy-to-understand explanations of predictions made by any ML model. Here, we give an overview of interpretability approaches using structured data and provide examples of practical interpretability of ML in different areas of healthcare, including prediction of healthrelated outcomes, optimizing treatments, or improving the efficiency of screening for specific conditions. Further, we outline future directions for interpretable ML and highlight the importance of developing algorithmic solutions that can enable ML driven decision making in high-stakes healthcare problems.

show abstract

The State of the Art in Integrating Machine Learning into Visual Analytics

Cited by 176 publications

References 162 publications

An Ontological Framework for Supporting the Design and Evaluation of Visual Analytics Systems

An Ontological Framework for Supporting the Design and Evaluation of Visual Analytics Systems

On the Challenges and Opportunities in Visualization for Machine Learning and Knowledge Extraction: A Research Agenda

Interpretability of machine learning‐based prediction models in healthcare

Contact Info

Product

Resources

About