Many dimensionality reduction algorithms have been proposed easing both tasks of visualization and classification in high dimension problems. Despite the different motivations they can be cast in a graph embedding fra...
详细信息
Many dimensionality reduction algorithms have been proposed easing both tasks of visualization and classification in high dimension problems. Despite the different motivations they can be cast in a graph embedding framework. In this paper we address weighted graph subspace learning methods for bankruptcy analysis. The rationale behind re-embedding the data in a lower dimensional space that would be better filled is twofold: to get the most compact representation (visualization) and to make subsequent processing of data more easy (classification). The approaches used, Graph regularized Non-Negative Matrix Factorization (GNMF) and Spatially Smooth Subspace Learning (SSSL), construct an affinity weight graph matrix to encode geometrical information and to learn in the training set the subspace models that enhance visualization and are able to ease the task of bankruptcy prediction. The experimental results on a real problem of French companies show that from the perspective of financial problem analysis the methodology is quite effective.
In this paper, a new data mining method is proposed on the basis of parallel coordinate for early warning of landslides. Landslides have resulted in many severe casualties and damaged structures and facilities. The pr...
详细信息
In this paper, a new data mining method is proposed on the basis of parallel coordinate for early warning of landslides. Landslides have resulted in many severe casualties and damaged structures and facilities. The proposed method is to analyse the landslide problems emerged with the parallel coordinates and its visualization function. It may simplify the establishment of complex model, and promote the visualization and analysis ability of spatial data, make closer relationship between spatial data and attribute data, and finally improve the effectiveness of landslide early warning.
The synergy between information visualization and knowledge visualization is explored using the "databaseTaxonomy" to guide the way. Drawing extensively on Burgin's mathematical theory of named sets, thi...
详细信息
The synergy between information visualization and knowledge visualization is explored using the "databaseTaxonomy" to guide the way. Drawing extensively on Burgin's mathematical theory of named sets, this new knowledge visualization treats all database content as ifit were scientific data, regardless of the database application. This mathematical tool penetrates deep into the logical structure of data and data relations, enabling a single algorithm to generate a conceptual knowledge structure, which pre-structures raw data in the database into a list of nested data-topic lists that works like a book index. For end-users, this visualization is familiar, convenient and precise. For the research community, this knowledge structure and the techniques used to build it offer an empirical tool for investigating the underlying properties of data, information, and knowledge on a computing device. The transformation from data to information to knowledge is automatic and seamless, thanks to a novel analysis of the logical structure of the symbols on these mechanical devices, one which reveals meta-symbols consisting of physical values(v) and constructed-types (t). The database Taxonomy also provides a first glimpse into how navigating this structure can generate a predicate logic expression, an outcome which the author believes promises to advance our theoretical understanding of knowledge visualization and of the influence of a digital media on symbolic logic.
Immunology produces large amounts of complex and hierarchical data. The overwhelming quantities and complexity of these data present a challenge for immunologists trying to interpret results, extract useful informatio...
详细信息
Immunology produces large amounts of complex and hierarchical data. The overwhelming quantities and complexity of these data present a challenge for immunologists trying to interpret results, extract useful information, and derive new knowledge. visualization plays an increasingly important role in the process of analyzing and understanding immunological data. We employed two visualization modules - heat map and stack graph, to be embedded in MULTIPRED2, a computational system for antigenic analysis and support of large-scale vaccine studies. We have described two complementary modules that display complex information on antigenic targets from proteins. The heat map enables visualization of large number of HLA variants, their groupings within the supertypes, and identification of antigenic regions within a query protein sequence. The stack graph is an interactive tool that presents predicted immunogenicity of a query protein across multiple HLA supertypes at human population level. The stack graph enables zooming in and out facilitating visualization at desired level of detail. Both visualization tools present a large amount of information in a single graphical display. The goal of visualization tools is to help immunologists and vaccine researchers to gain rapid insight into the data using comprehensive but clear graphic representation and summarization.
Nowadays, a great amount of data is created and distributed on the Internet. Tagging has become common practice to structure these data for easy access. Often the data and the associated tags contain spatial and tempo...
详细信息
Nowadays, a great amount of data is created and distributed on the Internet. Tagging has become common practice to structure these data for easy access. Often the data and the associated tags contain spatial and temporal information. In this paper, we develop general design strategies for visualizing spatially and temporally referenced tags similar to tag clouds on maps. Temporal information of tags is encoded through the visual appearance of text or through additional visual artifacts associated with the tags, whereas the location of tags on a map illustrates the spatial references. We demonstrate our solution based on an interactive visualization prototype for the exploration of both spatial and temporal references of Flickr tags.
We present Net EvViz, a visualization tool for analysis and exploration of a dynamic social network. There are plenty of visual social network analysis tools but few provide features for visualization of dynamically c...
详细信息
We present Net EvViz, a visualization tool for analysis and exploration of a dynamic social network. There are plenty of visual social network analysis tools but few provide features for visualization of dynamically changing networks featuring the addition or deletion of nodes or edges. Our tool extends the code base of the Node XL template for Microsoft Excel, a popular network visualization tool. The key features of this work are (1) The ability of the user to specify and edit temporal annotations to the network components in an Excel sheet, (2) See the dynamics of the network with multiple graph metrics plotted over the time span of the graph, called the Timeline, and (3) Temporal exploration of the network layout using an edge coloring scheme and a dynamic Time slider. The objectives of the new features presented in this paper are to let the data analysts, computer scientists and others to observe the dynamics or evolution in a network interactively. We presented Net EvViz to five users of Node XL and received positive responses.
Most commercial intrusion detections systems (IDS) can produce a very high volume of alerts, and are typically plagued by a high false positive rate. The approach described here uses Splunk to aggregate IDS alerts. Th...
详细信息
Most commercial intrusion detections systems (IDS) can produce a very high volume of alerts, and are typically plagued by a high false positive rate. The approach described here uses Splunk to aggregate IDS alerts. The aggregated IDS alerts are retrieved from Splunk programmatically and are then clustered using text analysis and visualized using a sunburst diagram to provide an additional understanding of the data. The equivalent of what the cluster analysis and visualization provides would require numerous detailed queries using Splunk and considerable manual effort.
To provide information efficiently to users with web service, we study information visualization techniques, especially focused on researcher network and graphic chart. In the viewpoint of data set and level of functi...
详细信息
To provide information efficiently to users with web service, we study information visualization techniques, especially focused on researcher network and graphic chart. In the viewpoint of data set and level of functionality, we analyze the following four academic information services, Authoratory, Research gate, Biomedexperts, and Academic. research. We analyze researcher network and graphic chart of each service and then propose evaluation criteria of visualization elements. We aim to contribute to practical development of technologies for information visualization that can enhance the usefulness of web information services.
The paper presents our experience in using sparse principal components (PCs) (Zou, Hastie and Tibshirani, 2006) for visualization of gearbox diagnostic data recorded for two bucket wheel excavators, one in bad and the...
详细信息
The paper presents our experience in using sparse principal components (PCs) (Zou, Hastie and Tibshirani, 2006) for visualization of gearbox diagnostic data recorded for two bucket wheel excavators, one in bad and the other in good state. The analyzed data had 15 basic variables. Our result is that two sparse PCs, based on 4 basic variables, yield similar display as classical pair of first two PCs using all fifteen basic variables. visualization of the data in Kohonen's SOMs confirms the conjecture that smaller number of variables reproduces quite well the overall structure of the data. Specificities of the applied sparse PCA method are discussed.
Extended Linguistic Dependency Diagrams are an innovative visualization of a data structure that is increasingly important in linguistics and language studies. It uses standard Info V is techniques in ways new to ling...
详细信息
Extended Linguistic Dependency Diagrams are an innovative visualization of a data structure that is increasingly important in linguistics and language studies. It uses standard Info V is techniques in ways new to linguistic diagrams to encode more information than is possible with previous visualizations. The goal is to make the diagrams easier to use, by allowing easier identification of the parts of the diagram of interest to the user. In addition, we aim to construct reusable tools to aid in language analysis and study. Preliminary evaluation supports the validity of the approach and suggests further improvements.
暂无评论