Deep Visual Analytics (DVA): Applications, Challenges and Future Directions

Islam, Rafiqul; Akter, Shanjita; Ratan, Rakybuzzaman; Kamal, Abu Raihan Mostofa; Xu, Guandong

doi:10.2991/hcis.k.210704.003

Cited by 13 publications

(7 citation statements)

References 109 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Leveraging the scalability and standardization of deep learning approaches can be helpful for VA. In this sense, the intersection between the two fields have evolved into Deep Visual Analytics (DVA) [2]. Normally, DVA systems apply deep learning to get predictive insights on the data, on top of other classic data visualisations.…”

Section: Backgrounds and Related Workmentioning

confidence: 99%

“…Both the purely exploratory tasks, such as finding cyclical patterns, anomalies, and clusters, as well as the predictive ones (classification, forecasting...) are, in most cases, carried out by visual analytics systems built ad-hoc, whose computational basis is based on statistics and KPIs that are not transferable between domains, and that are not scalable to big datasets. To solve these complex issues, Deep Visual Analytics (DVA) is starting to grow as an alternative way of designing and implementing Visual Interactive Systems (VISs) powered by neural networks [2].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DeepVATS: Deep Visual Analytics for Time Series

Rodríguez-Fernández¹,

Montalvo²,

Piccialli³

et al. 2023

Preprint

View full text Add to dashboard Cite

The field of Deep Visual Analytics (DVA) has recently arisen from the idea of developing Visual Interactive Systems supported by deep learning techniques, in order to provide them with large-scale data processing capabilities and to unify their implementation across different data modalities and domains of application. In this paper we present DeepVATS, an open-source tool that brings the field of DVA into time series data. DeepVATS trains, in a self-supervised way, a masked time series autoencoder that reconstructs patches of a time series, and projects the knowledge contained in the embeddings of that model in an interactive plot, from which time series patterns and anomalies emerge and can be easily spotted. The tool has been tested on both synthetic and real datasets, and its code is publicly available on https://github.com/vrodriguezf/deepvats

show abstract

Section: Backgrounds and Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

DeepVATS: Deep Visual Analytics for Time Series

Rodríguez-Fernández¹,

Montalvo²,

Piccialli³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…As new technologies continue to evolve in the field of visual analytics, it is crucial to evaluate their impact and effectiveness. 105 Further research can explore the utilization of established ES to assess the effectiveness of advanced technologies such as augmented reality, virtual reality, or natural language processing in visual analytics systems. Additionally, future studies can also investigate the possibility of combining and integrating diverse evaluation methodologies to enhance our understanding of these systems as shown in Figure 5.…”

Section: Future Directionsmentioning

confidence: 99%

Strategies for evaluating visual analytics systems: A systematic review and new perspectives

Islam,

Akter,

Islam

et al. 2023

Information Visualization

Self Cite

View full text Add to dashboard Cite

In recent times, visual analytics systems (VAS) have been used to solve various complex issues in diverse application domains. Nonetheless, an inherent drawback arises from the insufficient evaluation of VAS, resulting in occasional inaccuracies when it comes to analytical reasoning, information synthesis, and deriving insights from vast, ever-changing, ambiguous, and frequently contradictory data. Hence, the significance of implementing an appropriate evaluation methodology cannot be overstated, as it plays a pivotal role in enhancing the design and development of visualization systems. This paper assesses visualization systems by providing a systematic exploration of various evaluation strategies (ES). While several existing studies have examined some ES, the extent of comprehensive and systematic review for visualization research remains limited. In this work, we introduce seven state-of-the-art and widely recognized ES namely (1) dashboard comparison; (2) insight-based evaluation; (3) log data analysis; (4) Likert scales; (5) qualitative and quantitative analysis; (6) Nielsen’s heuristics; and (7) eye trackers. Moreover, it delves into their historical context and explores numerous applications where these ES have been employed, shedding light on the associated evaluation practices. Through our comprehensive review, we overview and analyze the predominant evaluation goals within the visualization community, elucidating their evolution and the inherent contrasts. Additionally, we identify the open challenges that arise with the emergence of new ES, while also highlighting the key themes gleaned from the existing literature that hold potential for further exploration in future studies.

show abstract

“…There are many challenges to providing an interactive visualization system, especially for the healthcare sector [19,20]. More recently, Rind et al [36] discussed that poor data quality and ambiguity are among the challenges in building visual analytics tools for temporal electronic health records.…”

Section: Visual Analytics For Explorationmentioning

confidence: 99%

“…However, traditional approaches are limited in their ability to consider such a large amount of data, where advanced technology such as visual analytics [19] provides the insightful impact to predict diabetes disease as well as identify the most significant factors. For example, Bhardwaj and Baliyan [7] visualized diabetics data in different ways and also proposed an interactive visualization system using Tableau.…”

Section: Introductionmentioning

confidence: 99%

DiaVis: Exploration and Analysis of Diabetes through Visual Interactive System

Rahman¹,

Islam²,

Akter³

et al. 2021

HCIS

Self Cite

View full text Add to dashboard Cite

Background: Diabetes is a long-term disease characterized by high blood sugar and has risen as a public health problem globally. Exploring and analyzing diabetes data is a timely concern because it may prompt a variety of serious illnesses, including stroke, kidney failure, heart attacks, etc. Several existing pieces of research have revealed that diabetes data, such as systolic blood pressure (SBP), diastolic blood pressure (DBP), weight, height, age, etc., can provide insightful information about patients diabetes states. However, very few studies have focused on visualizing diabetes mellitus (DM) insights to support healthcare administrator (HA)'s goals adequately, such as (i) decision-making, (ii) identifying and grouping associated factors, and (iii) analyzing large data effectively remains unexplored.Objective: This study aims to design an interactive visual system (Vis) to explore diabetes mellitus (DM) insights and its associated factors in Bangladesh. Methods:In this study, first, a case study method has employed to understand diabetes data. Second, we examine the potential of user-centered technology in addressing these challenges and design a Vis named "DiaVis" to process and present raw data in the form of graphics, graphs, and processed text, as well as a variety of user interaction possibilities. It helps to extract valuable data and present it in a simple and easy-to-understand way. Moreover, we highlight some key insights from our study that may help explore the healthcare community.Results: A user study with 20 individuals is used to evaluate our system. By allowing iterative exploration and modification of data in a dashboard with multiple-coordinated views, the DiaVis system improves the flow of visual analysis. Conclusion:This study suggests that the healthcare community should pay more attention to developing appropriate policy measures to reduce the risk of DM.

show abstract

Deep Visual Analytics (DVA): Applications, Challenges and Future Directions

Cited by 13 publications

References 109 publications

DeepVATS: Deep Visual Analytics for Time Series

DeepVATS: Deep Visual Analytics for Time Series

Strategies for evaluating visual analytics systems: A systematic review and new perspectives

DiaVis: Exploration and Analysis of Diabetes through Visual Interactive System

Contact Info

Product

Resources

About