Browsing by Author "Rasmussen, Matthew"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    A Graphical Interface to Clustering Algorithms and Visualizations
    (2004-05-25) Rasmussen, Matthew
    Due to recent advances in information technology, there has been an enormous growth in the amount of data generated in fields ranging from business to the sciences. This data has the potential to greatly benefit its owners by increasing their understanding of their own work. However, the growing size and complexity of data has introduced new challenges in extracting its meaning. To address these challenges, many data mining techniques have been developed. One technique in particular, clustering, has been successful in a wide range of applications. Clustering solves the general problem of identifying groups of related objects. Depending on the application, these objects may represent customers, documents, molecules, or genes. The ability to handle such diverse data in a general way has led to the popularity of clustering algorithms. In this paper, we introduce gCluto, a stand alone clustering software package designed to ease the use of clustering algorithms and their results. gCluto offers improvements over existing tools with features such as an intuitive graphical user interface, interactive visualizations, and mechanisms for comparing multiple clustering solutions. In addition to introducing the tool, the underlining algorithms and design decisions of gCluto will also be presented.
  • Thumbnail Image
    Item
    gCLUTO -- An Interactive Clustering, Visualization, and Analysis System
    (2004-05-25) Rasmussen, Matthew; Karypis, George
    Clustering algorithms are exploratory data analysis tools that have proved to be essential for gaining valuable insights on various aspects and relationships of the underlying systems. In this paper we present gCLUTO, a stand-alone clustering software package which serves as an easy-to-use platform that combines clustering algorithms along with a number of analysis, reporting, and visualization tools to aid in interactive exploration and clustering-driven analysis of large datasets. gCLUTO provides a wide-range of algorithms that operate either directly on the original feature-based representation of the objects or on the object-to-object similarity graphs and are capable of analyzing different types of datasets and finding clusters with different characteristics. In addition, gCLUTO implements a project-oriented work-flow that eases the process of data analysis.
  • Thumbnail Image
    Item
    gCluto: A graphical clustering toolkit
    (2003-12-16) Rasmussen, Matthew; Karypis, George
    gCLUTO (Graphical CLUstering TOolkit) is a graphical front-end for the CLUTO data clustering library. Its purpose is to make CLUTO's clustering abilities available in a user-friendly and graphical way. In addition, gCLUTO provides several ways to interactively visualize clustered results. A copy of gCLUTO can be found at http://www.cs.umn.edu/~mrasmus/gcluto. For more information about CLUTO visit http://www.cs.umn.edu/~karypis/cluto.
  • Thumbnail Image
    Item
    wCLUTO: A Web-Enabled Clustering Toolkit
    (2003-02-19) Rasmussen, Matthew; Deshpande, Mukund; Karypis, George; Johnson, James; Crow, John A.; Retzel, Ernest F.
    As structural and functional genomics efforts provide the biological community with ever-broadening sets of inter-related data, the need to explore such complex information for subtle relationships expands. We present wCluto, a web-enabled version of the stand-alone application Cluto, designed to apply clustering methods to genomic information.Its first application is focused on the clustering transcriptome data from microarrays. Data can be uploaded by the user into the clustering tool, a choice of several clustering methods can be made and configured, and data is presented to the user in a variety of visual formats,including a three-dimensional "mountain" view of the clusters. Parameters can be explored to rapidly examine a variety of clustering results, and the resulting clusters can be downloaded either for manipulation by other programs or saved in a format for publication.