Dimensionality reduction (DR) techniques for multidimensionaldata serve as powerful tools for visualization and understanding of the structure of the data. Various DR methods have been developed to extract specific f...
详细信息
Dimensionality reduction (DR) techniques for multidimensionaldata serve as powerful tools for visualization and understanding of the structure of the data. Various DR methods have been developed to extract specific features of the data over the years. However, selection of the optimal DR method and fine-tuning parameters are still challenging, as these choices vary based on the characteristics of the dataset. Consequently, data scientists often rely on their experience or undertake extensive experimentation to identify the most suitable approach. This paper proposes a semi-automatic method for selecting appropriate DR techniques through scatterplot evaluation. Initially, our approach applies a range of DR methods to the given multidimensionaldata to compute two-dimensional values. Next, we generate scatterplots from the two-dimensional data and calculate scores reflecting the distribution and spatial relationships among the points. Scatterplots that provide insights achieve higher scores, enabling an efficient selection of DR methods based on their visualization. We demonstrate the effectiveness of the presented method through two case studies: The first one is an e-commerce review dataset, and the second focuses on a dataset derived from music feature extraction.
This paper presents a user study of Variable Mapper, an interactive web-based geospatial visualization tool that aims to address the gap in multidimensional spatial data analysis. In the era of big data, researchers a...
详细信息
This paper presents a user study of Variable Mapper, an interactive web-based geospatial visualization tool that aims to address the gap in multidimensional spatial data analysis. In the era of big data, researchers and policymakers often grapple with complex, multi-faceted concepts situated within ‘wicked problems.’ These problems require the investigation of large numbers of interacting social and environmental variables and consider the data with different lenses. However, tools for visually exploring and representing such data remain inaccessible to researchers and policymakers with limited technical expertise. To address this gap, we describe and evaluate a new interactive tool for visualizing multidimensional geospatial data, which has been built in three stages: (1) development of a prototype through a case study of urban ‘superdiversity’, (2) refinement of the prototype by adding a case study of ‘liveability’, and (3) evaluation of the tool through a systematic user study. The paper explains the process of interdisciplinary collaboration in the development, implementation, and evaluation of the tool and discusses the results of the user study. Using mixed-method analysis, we generated six insights based on the user study: situating the data, locating multiple visualizations, making data apparent, comparing small or missing values, challenges in three dimensions, and managing cognitive load. These insights from the development, implementation, and evaluation of Variable Mapper contribute to the wider ongoing challenge of visualizing multidimensional geospatial data.
multidimensional data visualization is one of the most active research topics in information visualization since various information in our daily life forms multidimensionaldatasets. Scatterplot selection is an effec...
详细信息
multidimensional data visualization is one of the most active research topics in information visualization since various information in our daily life forms multidimensionaldatasets. Scatterplot selection is an effective approach to represent essential portions of multidimensionaldata in a limited display space. Various metrics for evaluating scatterplots, such as scagnostics, have been applied to scatterplot selection. One of the open problems of this research topic is that various scatterplots cannot be selected if we simply apply one of the metrics. In other words, we may want to apply multiple metrics simultaneously in a balanced manner when we want to select a variety of scatterplots. This paper presents a new scatterplot selection technique that solves this problem. First, the technique calculates the scores of scatterplots with multiple metrics and then constructs a graph by connecting pairs of scatterplots that have similar scores. Next, it uses a graph coloring algorithm to assign different colors to scatterplots that have similar scores. We can extract a set of various scatterplots by selecting them that the specific same color is assigned. This paper introduces two case studies: the former study is with a retail transaction dataset while the latter study is with a design optimization dataset.
"Skin Transparency" is an important keyword for women of all generations as one of the conditions for beautiful skin. Although, no one has clear definitions of "Skin Transparency". As it stands, be...
详细信息
ISBN:
(纸本)9781665438278
"Skin Transparency" is an important keyword for women of all generations as one of the conditions for beautiful skin. Although, no one has clear definitions of "Skin Transparency". As it stands, beauty experts are often invited to use visual methods in conducting skin transparency evaluation;however, it has not been still sufficiently clarified which visual properties are related to skin transparency. In this study, we aim to discover the relations between skin image features and sensory evaluations applying real human skin images. Specifically, we investigate "Skin Transparency" by comparing them using the Parallel Coordinate Plots. We observed their complex distributions by the visualization task.
In this paper, we investigate the use of information visualization techniques for getting insight into multidimensional financial data. In particular, we focus on exploring different multidimensionaldata visualizatio...
详细信息
ISBN:
(纸本)9781905305544
In this paper, we investigate the use of information visualization techniques for getting insight into multidimensional financial data. In particular, we focus on exploring different multidimensional data visualization techniques with respect to their effectiveness in solving a financial problem, namely financial competitor benchmarking. Financial competitor benchmarking is concerned with comparing the financial performance of different companies competing in the same market, industry, country or region. We investigate the extent to which different multidimensionalvisualization techniques are effective in revealing interesting patterns in financial performance data. For this purpose, we conducted a user evaluation study in which nine multidimensional data visualization techniques were assessed. The assessment concerns the extent to which users of these techniques are capable of discovering interesting patterns in multidimensional financial data, patterns associated with the problem of financial benchmarking. These patterns are identified as outlier detection, clustering, cluster description, class description and data comparison. The visualization techniques under analysis are: multiple line graphs, permutation matrix, survey plot, scatter plot matrix, parallel coordinates, treemap, principal components analysis, Sammon's mapping and the Self-Organizing Maps. The evaluation method consists in questionnaire-based data collection and analysis. We obtained answers from 12 students who agreed upon participating in this study. The evaluation we have conducted is useful especially in the early stage of the development of a visualization system, because it helps in the process of selection of most appropriate techniques for solving certain tasks.
When Analyzing multidimensional, quantitative data set, statistics of value distribution in each dimension, and comparison of two or more dimensions are common tasks in many domains, such as students' scores, coll...
详细信息
ISBN:
(纸本)9781457715846
When Analyzing multidimensional, quantitative data set, statistics of value distribution in each dimension, and comparison of two or more dimensions are common tasks in many domains, such as students' scores, college admissions and stocks. In this paper, we present a multidimensional data visualization method based on parallel coordinates and enhanced ring (PCER). The interactive scheme allows users to do a preliminary statistics on data to get an overview of data's distribution. Users can change the display proportion of one single dimension or all dimensions according to the preliminary statistical results and relative statistical requirement. The enhanced ring will show the distribution of the clutter area in parallel coordinates according to statistical results to help users getting more detailed information about data distribution and relationships among data. data are filtered by the interaction between parallel coordinates and enhanced ring according to user's statistical requirement to reduce the size of dataset and optimize the visualization effect. The experiment on students' scores shows that PCER is efficient in presenting and quantizing the distribution of students' scores.
visualization of unlabeled multidimensionaldata is commonly performed using projections to a 2D visual space, which supports an investigative interactive analysis. However, static views obtained by a projection metho...
详细信息
ISBN:
(纸本)9789897583063
visualization of unlabeled multidimensionaldata is commonly performed using projections to a 2D visual space, which supports an investigative interactive analysis. However, static views obtained by a projection method like Principal Component Analysis (PCA) may not capture well all data features. Moreover. in case of large data with many samples, the scatterplots suffer from overplotting, which hinders analysis purposes. Clustering tools allow for aggregation of data to meaningful structures. Clustering methods like K-means, however, also suffer from drawbacks. We present a novel approach to visually encode aggregated data in projected views and to interactively explore the data. We make use of the benefits of PCA and K-means clustering, but overcome their main drawbacks. The sensitivity of K-means to outlier points is ameliorated, while the sensitivity of PCA to axis scaling is converted into a powerful flexibility, allowing the user to change observation perspective by rescaling the original axes. Analysis of both clusters and outliers is facilitated. Properties of clusters are visually encoded in aggregated form using color and size or examined in detail via local scatterplots or local circular parallel coordinate plots. The granularity of the data aggregation process can be adjusted interactively. A star coordinate interaction widget allows for modifying the projection matrix. To convey how much the projection maintains neighborhoods, we use a distance encoding. We evaluate our tool using synthetic and real-world data sets and perform a user study to evaluate its effectiveness.
This paper presents an efficient visualization and exploration approach for modeling and characterizing the relationships and uncertainties in the context of a multidimensional ensemble dataset. Its core is a novel di...
详细信息
This paper presents an efficient visualization and exploration approach for modeling and characterizing the relationships and uncertainties in the context of a multidimensional ensemble dataset. Its core is a novel dissimilarity-preserving projection technique that characterizes not only the relationships among the mean values of the ensemble data objects but also the relationships among the distributions of ensemble members. This uncertainty-aware projection scheme leads to an improved understanding of the intrinsic structure in an ensemble dataset. The analysis of the ensemble dataset is further augmented by a suite of visual encoding and exploration tools. Experimental results on both artificial and real-world datasets demonstrate the effectiveness of our approach.
When Analyzing multidimensional, quantitative data set, statistics of value distribution in each dimension, and comparison of two or more dimensions are common tasks in many domains, such as students' scores, ...
详细信息
ISBN:
(纸本)9781457715860
When Analyzing multidimensional, quantitative data set, statistics of value distribution in each dimension, and comparison of two or more dimensions are common tasks in many domains, such as students' scores, college admissions and stocks. In this paper, we present a multidimensional data visualization method based on parallel coordinates and enhanced ring (PCER). The interactive scheme allows users to do a preliminary statistics on data to get an overview of data's distribution. Users can change the display proportion of one single dimension or all dimensions according to the preliminary statistical results and relative statistical requirement. The enhanced ring will show the distribution of the clutter area in parallel coordinates according to statistical results to help users getting more detailed information about data distribution and relationships among data. data are filtered by the interaction between parallel coordinates and enhanced ring according to user's statistical requirement to reduce the size of dataset and optimize the visualization effect. The experiment on students' scores shows that PCER is efficient in presenting and quantizing the distribution of students' scores.
The space-filling visualization model was first invented by Ben Shneiderman [28] for maximizing the utilization of display space in relational data (or graph) visualization, especially for tree visualization. It uses ...
详细信息
The space-filling visualization model was first invented by Ben Shneiderman [28] for maximizing the utilization of display space in relational data (or graph) visualization, especially for tree visualization. It uses the concept of Enclosure which dismisses the "edges" in the graphic representation that are all too frequently used in traditional node-link based graph visualizations. Therefore, the major issue in graph visualization which is the edge crossing can be naturally solved through the adoption of a space filling approach. However in the past, the space-filling concept has not attracted much attention from researchers in the field of multidimensionalvisualization. Although the problem of 'edge crossing' has also occurred among polylines which are used as the basic visual elements in the parallel coordinates visualization, it is problematic if those 'edge crossings' among polylines are not evenly distributed on the display plate as visual clutter will occur. This problem could significantly reduce the human readability in terms of reviewing a particular region of the visualization. In this study, we propose a new Space-Filling multidimensional data visualization (SFMDVis) that for the first-time introduces a space-filling approach into multidimensional data visualization. The main contributions are: (1) achieving the maximization of space utilization in multidimensionalvisualization (i.e. 100% of the display area is fully used), (2) eliminating visual clutter in SFMDVis through the use of the non-classic geometric primitive and (3) improving the quality of visualization for the visual perception of linear correlations among different variables as well as recognizing data patterns. To evaluate the quality of SFMDVis, we have conducted a usability study to measure the performance of SFMDVis in comparison with parallel coordinates and a scatterplot matrix for finding linear correlations and data patterns. The evaluation results have suggested that the accuracy
暂无评论