Wise, J.A., Thomas, J.J., Pennock, K., Lantrip, D., Pottier, M., Schur, A., Crow, V. Visualizing the nonvisual: Spatial analysis and interaction with information from text documents. IEEE Information Visualization 95, 1995  Home/Search   Document Not in Database   Summary   Related Articles   Check

....user s interests. For instance, if the user has been reading about oil exploration in Alaska, a top level split labeled Texaco, Exxon, Mobil might be the way he she mentally organizes the information space, but there are many other perspectives for which this organization does not help. Themeview [6] is perhaps the best known text document corpus visualization. Like most text information retrieval systems, it treats a document as a vector of normalized word counts, with one element for each word in the corpus vocabulary. This is usually at least several thousand. In order to visualize this ....

Wise, J.A., J. Thomas, J., K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents, in IEEE Information Visualization. 1995. p. 51-58.

....the layout and rendering of large graphs in the context of information visualization can be found in [4, 3] Landscape visualization techniques have been used for text document visualization. Wise et al. developed the ThemeScape technique, which conveys information about topics in text documents, [13]. ThemeScapes are abstract, three dimensional landscapes of information that are constructed from document corpora which augment a 2D landscape of text with a height dimension showing the strength of a theme in a given region. Elevation depicts theme strength, while other features of the terrain ....

J.A. Wise, J.J. Thomas, K. Pennock, and D. Lantrip. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. In S.K. Card, J.D. Mackinlay, and B. Shneiderman, editors, Readings in Information Visualization, pages 442--450. Morgan Kaufmann Publishers, 1999.

....in a straightforward manner, by linking point symbols and labels to the computed point locations. Point configurations are useful for the creation of landscape visualizations. Feature attributes can be linked to point locations as elevations and interpolated to form a continuous surface [21]. While this can result in a visually attractive representation, the attribute to be interpolated as well as the interpolation function and parameters have to be chosen carefully. Existing proposals in this direction rarely consider how meaningful the mixture of continuous surfaces with discrete ....

J. A. Wise, J. J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow, "Visualizing the Non-Visual: Spatial Analysis and Interaction with Information from Text Documents," presented at InfoVis '95, Atlanta GA, 1995.

....data. Different variables can be plotted against each other to allow analysts to identify relationships between them, facilitating data mining. These techniques can be combined with those of information retrieval to produce visualisations of document collections, as in the Bead [15, 21] and SPIRE [126] systems. Text documents have no obvious coordinates for convenient representation on a screen, and so these systems attempt to place points in a two or three dimensional space such that their relative proximity reflects the similarity of the corresponding documents. Usually, clusters are not ....

....further in Appendix A. Basalaj [7] implemented a number of MDS algorithms and compared them to related techniques such as principal components analysis (PCA) finding that visualisations created with MDS were generally quantitatively and qualitatively superior. As Wise and his colleagues [126] have noted, a visualisation may help peo ple to make sense of a large collection of documents while having to read fewer of them, thereby ameliorating the problem of information overload. In this research, however, we were primarily interested in the use of visualisation techniques as a means of ....

J. A. Wise, J. J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. In Proceedings of the IEEE Symposium on Information Visualization (InfoVis'95), pages 51--58. IEEE, 1995.

....preserving as much as possible the structure of the data in the high dimensional data space. This is achieved by mapping points in one space to points in another space such that nearby points map to nearby points (and sometimes in addition far away points map to far away points) 11] Galaxies [32, 31] visualization displays clusters and document interrelatedness by reducing a high dimensional representation of documents to a two dimensional scatterplot. The documents are clustered in the high dimensional space through a metric of similarity such as Euclidean distance or cosine measures. Then ....

J. A. Wise, J. J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. Proc. of Information Visualization '1995.

.... and display of the original full text of its article at ACM s electronic proceedings website for detailed study [7] Researchers at the Pacific Northwest National Laboratory have developed a suite of new technologies called SPIRE TM, or Spatial Paradigm for Information Retrieval and Exploration [22]. SPIRE accepts large volumes of unformatted text, determines the dominant topics and relationships within the text, and graphically displays them based on word similarities and themes in text such that similar texts are constrained to lie close to one another. The visualizations allow the user to ....

Wise, J. & Thomas, J., "Visualizing the Non-Visual: Spatial Analysis and Interaction with Information from Text Documents," In Proceedings of IEEE Information Visualization, pages 51-58. IEEE Service Center, Atlanta, GA, October 1995.

....system: Most of the current software systems for graphically presenting data are not built to handle a wide range of data types and hence they tend to be special purpose. For example, Parallel Coordinates [13] for numeric variables, Mosaic Displays [7] for nominal data, ThemeView and Galaxies [36] for text corpus are few exemplars of many such techniques. The reason of the tools being special purpose could be that the data that the visualization system uses to generate graphics is primarily numeric. This is because numeric data can be easily mapped to graphical attributes, which are ....

....are many text visualization tools available. Bead [4] presents similarity relationships between documents by 3dimensional particle clouds. Using a similar visualization approach, Starlight [28] aims to visualize the content of multimedia databases. Closely related to these approaches is Galaxies [36] which yields a simple 2D scatter plot. VxInsight [6] visualizes the document distribution density by means of a mountain terrain metaphor. These approaches are focused on the display of similarity between individual documents. The scaling techniques used try to optimize the distances between ....

J. A. Wise, J. J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. In Proc. of Information Visualization '1995.

....the right images more easily and intuitively. The advances of information visualization and data mining techniques now allow users to explore an 0 7695 0743 3 00 10.00 2000 IEEE information space organized through a variety of metaphors, such as an information landscape or an information galaxy [7, 8]. Many of these visualizations are based on interrelationships derived from textual information, typically using classic information retrieval models such as the vector space model [9] Latent Semantic Indexing (LSI) 10] or other variants. There has been a steadily increased interest in a ....

J.A. Wise Jr., J. J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow, "Visualizing the non-visual: Spatial analysis and interaction with information from text documents," Proceedings of IEEE Symposium on Information Visualization '95, Atlanta, Georgia, USA, 1995.

....data down to a reduced number of dimensions. Kohonen s Self Organizing Maps (SOM) 16, 6] is an unsupervised learning method for reducing multidimensional data to 2D feature maps [3] There are many visualization systems that make use of existing dimensionality reduction techniques [23, 3, 11]. Galaxies and ThemeScape [23] project high dimensional document vectors and their cluster centroids down into a two dimensional space, and then use scatterplots and landscapes to visualize them [22] Bead [3] uses MDS to lay out high dimensional data in a two or three dimensional space and uses ....

....number of dimensions. Kohonen s Self Organizing Maps (SOM) 16, 6] is an unsupervised learning method for reducing multidimensional data to 2D feature maps [3] There are many visualization systems that make use of existing dimensionality reduction techniques [23, 3, 11] Galaxies and ThemeScape [23] project high dimensional document vectors and their cluster centroids down into a two dimensional space, and then use scatterplots and landscapes to visualize them [22] Bead [3] uses MDS to lay out high dimensional data in a two or three dimensional space and uses imageability features to ....

J. A. Wise, J. J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. Proc. of Information Visualization '1995.

....that can extract relevant segments from unstructured text and organize them in a way that one can easily retrieve them. 3.3. ThemeView SPIRE, the Spatial Paradigm for Information Retrieval and Exploration is a classic example of information visualization developed by Pacific Northwest Laboratory [38]. SPIRE in fact is a suite of visualization tools for browsing and selecting text documents from large corpora, including a visualization view called Themescapes, which is later known as ThemeView. The SPIRE project is funded by the Department of Energy and the U.S. intelligence agencies. ....

Wise, J.A., Thomas, J.J., Pennock, K., Lantrip, D., Pottier, M., Schur, A. and Crow, V., Visualizing the non-visual: Spatial analysis and interaction with information from text documents. in IEEE Symposium on Information Visualization '95, (Atlanta, Georgia, USA, 1995), IEEE Computer Society Press.

....so that improvements can be made. Strategists would also like to mine information about users, such as product interest. Users need tools to navigate and locate information faster. To support these 55 sensemaking tasks, visualization of large hypertext spaces has been done by various researchers [5, 7, 27, 46, 109, 78, 69]. These systems are designed with a fixed priori with a limited set of tasks in mind. A system has yet to provide a set of primitives for conducting iterative and cyclic analysis tasks on Web ecologies. In our work, we provide such a system by handling very large Web sites (15,000 files) and ....

James A. Wise, James J. Thomas, Kelly Pennock, David Lantrip, Marc Pottier, Anne Schur, and Vern Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. In Proceedings of the Symposium on Information Visualization '95, pages 51--58. IEEE CS, 1995. Atlanta, Georgia.

....in different colors. Various combinations of characters can be rendered together by means of the slider. Fig. 11. SeeSoft[18] Table 13. SeeSoft (See Fig. 11) Name D F D X Y Z T R [ CP Line# Q Q P Type O O C Type O sl O C Another mapping of text is represented in Themescapes [19]. The text for each document (for example, a news story) is transformed into a document vector. Document vectors are compared giving rise to a matrix of similarities. The matrix is mapped onto a 2D landscape by means of multi dimensional scaling. A mds X:P, Y:P This gives a 2D map of ....

....surface. Q Z:S . The result is shown in Fig. 12, which depicts as a landscape, themes from CNN news. Thus in text, as in other specialized data areas, the transformations from the raw data type to a visualizable data type can be as important as the actual visualization. Fig. 12. Themescapes[19] Table 14. Themescapes Name D F V X Y Z T R [ A Content A mds xy P P Number Q S C SUMMARY In this paper we have sketched part of a scheme for mapping the morphology of the design space of visualizations. Because of space limitations, we have only sampled from the set of techniques being ....

J. A. Wise, J. J. Thomas, K. Pennock, D. Lantrip, M. Pottier, and A. Schur, "Visualizing the non-visual: Spatial analysis and interaction with information from text documents," in InfoVis '95. New York: ACM, 1995.

....A 3D node link environment with node aggregation. 95 3.13 The CHEOPS node compression method . 96 3.14TwotypesofFisheyeviews. 97 3.15TheBeadslandscapeasdescribedin[24] 99 3. 16 The original Themescape as described in [172]. 100 3.17 NicheWorks constellation . 101 3.18 Highlighting of multiple intersections in Attribute Explorer. 103 3.19TheoriginalMagicLens . 104 3.20Movablelensfilters. 104 3.21 ....

....3 terms. Lyberworld uses a 3D sphere as the boundary of space to further minimise the problem of interpretation, allowing a higher number of terms to be used. VRVIBE also encouraged multiple participates in the 3D environment to browse the content together. Beads [33] VxInsight [46] Themescape [172] (and its commercial incarnation at Cartia, Inc. 29] and Nicheworks [171] are systems that position document nodes according to their mutual and global attraction. Beads has evolved over a number of years and as such includes many interesting novel techniques: starting from triangular nodes in ....

[Article contains additional citation context not shown here]

WISE, J. A., THOMAS,J.J.,PENNOCK, K., LANTRIP, D., POTTIER,M., SCHUR, A., AND CROW, V. Visualizing the Non-Visual: Spatial Analysis and Interaction with Information from Text Documents. In Proceedings of InfoViz'95, IEEE Symposium on Information Visualization (30--31 October 1995), IEEE Computer Society Press, pp. 51--58.

....a modular framework for the computation of document maps. 2. RELATED WORK In literature two groups of document map approaches can be found. The first uses Multidimensional Scaling or related techniques for calculating a document space and generates scatter plots for visualization (cf. 4] 13] [18]) The scaling techniques used try to optimize the distances between the documents with respect to the given proximity measure. The corresponding systems display relationships of individual documents. The second group of approaches uses a self organizing feature map (SOFM) as a basis for ....

Wise, J.A., Thomas, J.J., Pennock, K., Lantrip, D., Pottier, M., Schur, A., Crow, V. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. IEEE Information Visualization 95, 1995

....algorithms made this visualization technique impractical for large collections of data. The authors report that it took 150 minutes to lay out 301 articles. Clearly several orders of magnitude of improvement in computing time are required to make this interface useful for interactive tasks. Wise et al. (1995) described interfaces for displaying two and three dimensional similarity based clusters of documents. They visualized a collection of transcripts from CNN news stories from one week. They did not report, however, which algorithms they used, how many documents they were displaying, or how ....

Wise, J.A., Thomas, J.J., Pennock, K., Lantrip, D., Pottier, M., Schur, A., and Crow, V. (1995) Visualizing the Non-Visual: Spatial Analysis and Interaction with Information from Text Documents. In Proceedings of Information Visualization '95, Atlanta, GA. IEEE Press. pp. 51-58.

....directly, e.g. using a scatter plot: BEAD [3] presents similarity relationships between documents by 3 dimensional particle clouds. Using a similar visualization approach, STARLIGHT [11] aims to visualize the content of multimedia databases. Closely related to these approaches is GALAXIES [15] which yields a simple 2D scatter plot. VXINSIGHT [5] visualizes the document distribution density by means of a mountain terrain metaphor. These approaches are focused on the display of similarity between individual documents. The scaling techniques used try to optimize the distances between ....

Wise, J.A., Thomas, J.J., Pennock, K., Lantrip, D., Pottier, M., Schur, A., Crow, V. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. In Proc. of IEEE Information Visualization 95 (InfoViz'95), 1995, 51--58

No context found.

Wise, J.A., Thomas, J.J., Pennock, K., Lantrip, D., Pottier, M., Schur, A., Crow, V. Visualizing the nonvisual: Spatial analysis and interaction with information from text documents. IEEE Information Visualization 95, 1995

No context found.

Wise, James A., Thomas, James J., Pen-nock, Kelly, Lantrip, David, Pottier, Marc, and Schur, Anne. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. In Proc. of the Information Visualization Symposium, pp. 51-58. IEEE Computer Society Press, 1995.

No context found.

J. Wise, et al.: Visualizing the Non-Visual: Spatial analysis and Interaction with Information from text Documents, Proceedings of the IEEE Info. Vis. 95, pp. 51-58, 1995

No context found.

J.A. Wise, J.J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. Proc. IEEE Symposium on Information Visualization, pages 51--58, 1995.

No context found.

J.A. Wise, J.J. Thomas, K. Pennock, D. Lantrip, M. Pottier, A. Schur, and V. Crow. Visualizing the non-visual: Spatial analysis and interaction with information from text documents. Proc. IEEE Symposium on Information Visualization, pages 51--58, 1995.

No context found.

J.A. Wise et al., "Visualizing the Nonvisual: Spatial Analysis and Interaction with Information from Text Documents, " Proc. of Information Visualization 95, IEEE Computer Soc. Press, Los Alamitos, Calif., 1995, pp. 51-58.

No context found.

Wise, J.A., Thomas, J.J., Permock, K., Lantrip, D., Pottier, M., Schur, A., & Crow, V. "Visualizing the Non-Visual: Spatial Analysis and Interaction with Information from Text Documents", Proceedings of lnfoVis '95, IEEE, 1995, 51-58.

No context found.

Wise, J. A., Thomas, J. J., Pennock, K., Lantrip, D., Pottier, M., Schur, A., and Crow, V., "Visualizing the non-visual: Spatial analysis and interaction with information from text documents", in Proceedings of Information Visualization Symposium (InfoVis'95), 1995, pp. 51-58.

No context found.

--355. Wise, J.A., et al. (1995). Visualizing the non-visual: Spatial analysis and interaction with information from text documents. In: N. Gershon & S. Eick (Eds.), Proceedings: Information visualization; 1995, October 30 --

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC