Research

18 pages, 9937 KiB

Open AccessArticle

Datamator: An Authoring Tool for Creating Datamations via Data Query Decomposition

by Yi Guo, Nan Cao, Ligan Cai, Yanqiu Wu, Daniel Weiskopf, Danqing Shi and Qing Chen

Appl. Sci. 2023, 13(17), 9709; https://doi.org/10.3390/app13179709 - 28 Aug 2023

Cited by 1 | Viewed by 735

Datamation is designed to animate an analysis pipeline step by step, serving as an intuitive and efficient method for interpreting data analysis outcomes and facilitating easy sharing with others. However, the creation of a datamation is a difficult task that demands expertise in [...] Read more.

Datamation is designed to animate an analysis pipeline step by step, serving as an intuitive and efficient method for interpreting data analysis outcomes and facilitating easy sharing with others. However, the creation of a datamation is a difficult task that demands expertise in diverse skills. To simplify this task, we introduce Datamator, a language-oriented authoring tool developed to support datamation generation. In this system, we develop a data query analyzer that enables users to generate an initial datamation effortlessly by inputting a data question in natural language. Then, the datamation is displayed in an interactive editor that affords users the ability to both edit the analysis progression and delve into the specifics of each step undertaken. Notably, the Datamator incorporates a novel calibration network that is able to optimize the outputs of the query decomposition network using a small amount of user feedback. To demonstrate the effectiveness of Datamator, we conduct a series of evaluations including performance validation, a controlled user study, and expert interviews. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

19 pages, 4242 KiB

Open AccessArticle

Visualization and Data Analysis of Multi-Factors for the Scientific Research Training of Graduate Students

by Yanan Liu, Guojun Li, Yulong Yin and Leibao Zhang

Appl. Sci. 2022, 12(24), 12845; https://doi.org/10.3390/app122412845 - 14 Dec 2022

Viewed by 1739

Abstract

With the change of graduate education from quantity expansion to quality promotion, how to improve the quality of graduate cultivation has aroused wide concern. However, existing scientific quantitative methods tend to investigate the results of graduate training, with a lack of attention to [...] Read more.

With the change of graduate education from quantity expansion to quality promotion, how to improve the quality of graduate cultivation has aroused wide concern. However, existing scientific quantitative methods tend to investigate the results of graduate training, with a lack of attention to the multidimensional data during the training process. Thus, exploratory analysis of multidimensional data in the graduate training process and accurate grasp of the key process factors affecting graduate academic competence is an indispensable task for achieving the stated goals of graduate education. In this paper, a visual analytic system of graduate training data is proposed to help users implement in-depth analysis based on the graduate training process. First, a questionnaire is designed about the training process to identify multidimensional data timely and accurately. Then, a series of data mining methods are utilized to further detect key factors in the training process, which will be used to make academic predictions for first-year graduates. Meanwhile, an interactive visual analytic system has been developed to help users understand and analyze the key factors affecting the graduate training process. Based on the results of the visual analysis, effective suggestions will be provided for graduate students, supervisors, and university administrators to improve the quality of graduate education. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

17 pages, 41502 KiB

Open AccessArticle

A Military Object Detection Model of UAV Reconnaissance Image and Feature Visualization

by Huanhua Liu, Yonghao Yu, Shengzong Liu and Wei Wang

Appl. Sci. 2022, 12(23), 12236; https://doi.org/10.3390/app122312236 - 29 Nov 2022

Cited by 16 | Viewed by 4703

Abstract

Military object detection from Unmanned Aerial Vehicle (UAV) reconnaissance images faces challenges, including lack of image data, images with poor quality, and small objects. In this work, we simulate UAV low-altitude reconnaissance and construct the UAV reconnaissance image tank database UAVT-3. Then, we [...] Read more.

Military object detection from Unmanned Aerial Vehicle (UAV) reconnaissance images faces challenges, including lack of image data, images with poor quality, and small objects. In this work, we simulate UAV low-altitude reconnaissance and construct the UAV reconnaissance image tank database UAVT-3. Then, we improve YOLOv5 and propose UAVT-YOLOv5 for object detection of UAV images. First, data augmentation of blurred images is introduced to improve the accuracy of fog and motion-blurred images. Secondly, a large-scale feature map together with multi-scale feedback is added to improve the recognition ability of small objects. Thirdly, we optimize the loss function by increasing the loss penalty of small objects and classes with fewer samples. Finally, the anchor boxes are optimized by clustering the ground truth object box of UAVT-3. The feature visualization technique Class Action Mapping (CAM) is introduced to explore the mechanisms of the proposed model. The experimental results of the improved model evaluated on UAVT-3 show that the mAP reaches 99.2%, an increase of 2.1% compared with YOLOv5, the detection speed is 40 frames per second, and data augmentation of blurred images yields an mAP increase of 20.4% and 26.6% for fog and motion blur images detection. The class action maps show the discriminant region of the tanks is the turret for UAVT-YOLOv5. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

19 pages, 7972 KiB

Open AccessArticle

Multivariate Fence: Using Parallel Coordinates to Locate and Compare Attributes of Adjacency Matrix Nodes in Immersive Environment

by Tiemeng Li, Yanning Jin, Songqian Wu and Shiran Liu

Appl. Sci. 2022, 12(23), 12182; https://doi.org/10.3390/app122312182 - 28 Nov 2022

Cited by 1 | Viewed by 2035

Abstract

Adjacency matrix visualization is a common method for presenting graph data, and the Focus+Context technique can be used to explore the details of the ROI (region of interest). Embedded views and multi-view approaches are usually applied when locating and comparing attributes among multiple [...] Read more.

Adjacency matrix visualization is a common method for presenting graph data, and the Focus+Context technique can be used to explore the details of the ROI (region of interest). Embedded views and multi-view approaches are usually applied when locating and comparing attributes among multiple nodes. However, the embedded view has an issue of edge occlusion, while the multi-view would cause repeated perspective switching. In this paper, we propose a Multivariate Fence (MVF) model as a focus view of the adjacency matrix to locate and compare attributes among nodes. An additional spatial parallel coordinate is added to the 2D adjacency matrix in an immersive environment so that the attribute information can be shown in a single view without blocking edge information. We also conduct a user study to evaluate the performance of the MVF. The results show that the MVF has better efficiency and accuracy in locating and comparing the multivariate adjacency matrix in the immersive environment against the existing focus model. Moreover, the MVF model is easier to understand and is preferred by users. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

19 pages, 1257 KiB

Open AccessArticle

Active Pattern Classification for Automatic Visual Exploration of Multi-Dimensional Data

by Jie Li, Huailian Tan and Wentao Huang

Appl. Sci. 2022, 12(22), 11386; https://doi.org/10.3390/app122211386 - 10 Nov 2022

Viewed by 1175

Abstract

The practice of applying a classifier (called a pattern classifier and abbreviated as PC below) in a visual analysis system to identify patterns from interactively generated visualizations is gradually emerging. Demonstrated cases in existing works focus on ideal scenarios where the analyst can [...] Read more.

The practice of applying a classifier (called a pattern classifier and abbreviated as PC below) in a visual analysis system to identify patterns from interactively generated visualizations is gradually emerging. Demonstrated cases in existing works focus on ideal scenarios where the analyst can determine all the pattern types in advance without adjusting the classifier settings during the exploration process. However, in most real-world scenarios, analysts know nothing about data patterns before exploring the dataset and inevitably find novel patterns during the exploration. This difference makes the traditional classifier training and application mode less suitable. Analysts have to artificially determine whether each generated visualization contains new data patterns to adjust the classifier setting, thus affecting the automation of the data exploration. This paper proposes a novel PC-based data exploration approach. The core of the approach is an active-learning indicator for automatically identifying visualizations involving new pattern classes. Analysts thus can apply PCs to explore data while dynamically adjusting the PCs using these visualizations. We further propose a PC-based visualization framework that takes full advantage of the PC in terms of efficiency by allowing analysts to explore an exploring space, rather than a single visualization at a time. The results of the quantitative experiment and the performance of participants in the user study demonstrate the effectiveness and usability of the method. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

26 pages, 11395 KiB

Open AccessArticle

VAPPD: Visual Analysis of Protein Pocket Dynamics

by Dongliang Guo, Li Feng, Chuanbao Shi, Lina Cao, Yu Li, Yanfen Wang and Ximing Xu

Appl. Sci. 2022, 12(20), 10465; https://doi.org/10.3390/app122010465 - 17 Oct 2022

Viewed by 1540

Abstract

Analyzing the intrinsic dynamic characteristics of protein pockets is a key aspect to understanding the functional mechanism of proteins, which is conducive to the discovery and development of drugs. At present, the research on the dynamic characteristics of pockets mainly focuses on pocket [...] Read more.

Analyzing the intrinsic dynamic characteristics of protein pockets is a key aspect to understanding the functional mechanism of proteins, which is conducive to the discovery and development of drugs. At present, the research on the dynamic characteristics of pockets mainly focuses on pocket stability, similarity, and physicochemical properties. However, due to the high complexity and diversity of high-dimensional pocket data in dynamic processes, this work is challenging. In this paper, we explore the dynamic characteristics of protein pockets based on molecular dynamics (MD) simulation trajectories. First, a dynamic pocket shape representation method combining topological feature data is proposed to improve the accuracy of pocket similarity calculation. Secondly, a novel high-dimensional pocket similarity calculation method based on pocket to vector dynamic time warp (P2V-DTW) is proposed to solve the correlation calculation problem of unequal length sequences. Thirdly, a visual analysis system of protein dynamics (VAPPD) is proposed to help experts study the characteristics of high-dimensional dynamic pockets in detail. Finally, the efficiency of our approach is demonstrated in case studies of GPX4 and ACE2. By observing the characteristic changes of pockets under different spatiotemporal scales, especially the motion correlation between pockets, we can find the allosteric pockets. Experts in the field of biomolecules who cooperated with us confirm that our method is efficient and reliable, and has potential for high-dimensional dynamic pocket data analysis. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

21 pages, 6721 KiB

Open AccessArticle

PRRGNVis: Multi-Level Visual Analysis of Comparison for Predicted Results of Recurrent Geometric Network

by Yanfen Wang, Li Feng, Quan Wang, Yang Xu and Dongliang Guo

Appl. Sci. 2022, 12(17), 8465; https://doi.org/10.3390/app12178465 - 24 Aug 2022

Viewed by 963

Abstract

The structure of a protein determines its function, and the advancement of machine learning has led to the rapid development of protein structure prediction. Protein structure comparison is crucial for inferring the evolutionary relationship of proteins, drug discovery, and protein design. In this [...] Read more.

The structure of a protein determines its function, and the advancement of machine learning has led to the rapid development of protein structure prediction. Protein structure comparison is crucial for inferring the evolutionary relationship of proteins, drug discovery, and protein design. In this paper, we propose a multi-level visual analysis method to improve the protein structure comparison between predicted and actual structures. Our method takes the predicted results of the Recurrent Geometric Network (RGN) as the main research object and is mainly designed following three levels of protein structure visualization on RGN. Firstly, at the prediction accuracy level of the RGN, we use the Global Distance Test—Total Score (GDT_TS) as the evaluation standard, then compare it with distance-based root mean square deviation (dRMSD) and Template Modeling Score (TM-Score) to analyze the prediction characteristics of the RGN. Secondly, the distance deviation, torsion angle, and other attributes are used to analyze the difference between the predicted structure and the actual structure at the structural similarity level. Next, at the structural stability level, the Ramachandran Plot and PictorialBar combine to be improved to detect the quality of the predicted structure and analyze whether the amino acid residues conform to the theoretical configuration. Finally, we interactively analyze the characteristics of the RGN with the above visualization effects and give reasons and reasonable suggestions. By case studies, we demonstrate that our method is effective and can also be used to analyze other predictive network results. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

17 pages, 2741 KiB

Open AccessArticle

MULTI-NETVIS: Visual Analytics for Multivariate Network

by Song Wang, Shijie Chen, Ting Cai and Yadong Wu

Appl. Sci. 2022, 12(17), 8405; https://doi.org/10.3390/app12178405 - 23 Aug 2022

Cited by 2 | Viewed by 1305

Abstract

In the process of multivariate network exploration and analysis, it is important to consider network topology and attribute correlation analysis. In order to consider both in presentation and analysis, existing research focuses on visual design and multiple-view visualization. However, this multitudinous visual design [...] Read more.

In the process of multivariate network exploration and analysis, it is important to consider network topology and attribute correlation analysis. In order to consider both in presentation and analysis, existing research focuses on visual design and multiple-view visualization. However, this multitudinous visual design makes network cognition and analysis difficult. In multi-view visualization, the associated information among attributes is rarely retained and is often accompanied by tedious interaction processes. In this paper, a layout scheme is proposed to balance attribute and topology analysis in multivariate network visual analysis and a multivariate network visual analytics system is implemented based on the layout scheme. The analysis scenarios of overall, community, and local multi-granularity are provided by the layout scheme, which combines 3D, 2.5D, and 2D layouts. According to the layout scheme, we propose a layout transformation method to maintain the relative position of the topological context layouts in three dimensions. Furthermore, we propose a Louvain-3D FDA layout algorithm for the 3D layout, and introduce an edge bundling algorithm in the 2.5D layout to achieve an attribute-oriented topology layout. Combining the principle of interaction from global to detail, we design a novel system, Multi-NetVis, which supports users in drilling exploration and analysis and takes both the network attribute correlations and topological structure into consideration. Finally, two datasets are selected to demonstrate the usage scenarios and an evaluation experiment is designed to verify the effectiveness of the layout scheme. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

20 pages, 6359 KiB

Open AccessArticle

PNMAVis: Visual Analysis Tool of Protein Normal Mode for Understanding Cavity Dynamics

by Dongliang Guo, Li Feng, Taoxiang Zhang, Yaoyao Guo, Yanfen Wang and Ximing Xu

Appl. Sci. 2022, 12(15), 7919; https://doi.org/10.3390/app12157919 - 07 Aug 2022

Viewed by 1523

Abstract

Molecular cavities play a critical role in our understanding of molecular phenomena. Recently, a number of works on the visual analysis of protein cavity dynamics have been developed to allow experts and users to interactively research dynamic cavity data. However, previous explorations are [...] Read more.

Molecular cavities play a critical role in our understanding of molecular phenomena. Recently, a number of works on the visual analysis of protein cavity dynamics have been developed to allow experts and users to interactively research dynamic cavity data. However, previous explorations are limited to studying cavity-lining amino acids and they lack a consideration of the impact of the key amino acids, which are far away from the cavity but have an important impact on the cavity. When studying protein amino acids, biochemists use normal mode decomposition to analyze protein changes on a time scale. However, the high-dimensional parameter space generated via decomposition is too large to be analyzed in detail. We present a novel approach that combines cavity characterization and normal mode analysis (NMA) for cavity dynamics analysis to reduce and explore this vast space through interactive visualization. PNMAVis can analyze whether direct factors (cavity-lining amino acids) or indirect factors (key amino acids) affect cavity changes, through multiple linked 2D and 3D views. The visual analysis method we proposed is based on close cooperation with domain experts, aiming to meet their needs to explore the relationship between cavity stability and cavity-lining amino acids fluctuations and key amino acids fluctuations as much as possible, and also to help domain experts identify potential allosteric residues. The effectiveness of our new method is demonstrated by the case study conducted by cooperative protein experts on a biological field case and an open normal mode data set. Full article

(This article belongs to the Special Issue Multidimensional Data Visualization: Methods and Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Multidimensional Data Visualization: Methods and Applications

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (9 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI