Welcome to MapForums!

Register, sign in, or use Facebook Connect above to join in and participate in the forum.

When you are signed in, this message, the ads in this row, and the red-underlined link ads all go away.

Subscribe to receive our newsletter.
Subscribe Unsubscribe
Results 1 to 1 of 1

Using MapPoint for Cluster Analysis: Part 1, Applications

This is a discussion on Using MapPoint for Cluster Analysis: Part 1, Applications within the MP2K Magazine Articles forums, part of the Map Forums category; Cluster analysis is a powerful geospatial tool that is used to find groups of data points. In a general sense, ...

  1. #1
    Richard Marsden is offline Junior Member White Belt
    Join Date
    Jun 2011
    Posts
    5

    Using MapPoint for Cluster Analysis: Part 1, Applications

    Cluster analysis is a powerful geospatial tool that is used to find groups of data points.

    In a general sense, cluster analysis is defined as the task of grouping a set of objects into groups with similar characteristics. The concept of ‘similarity’ is not very well defined, hence a wide range of algorithms have been developed for the wide range of potential applications and their characteristics. For geospatial applications, the shared characteristic is that of location, although other characteristics (e.g. sales volumes) can be used to restrict the size of clusters.



    An early pioneering use of geospatial cluster analysis was by John Snow in 1854 when he plotted a map of the 1854 Broad Street Cholera Outbreak in London. By plotting each case of cholera on a map, John Snow was able to identify that the outbreak was centered on a specific water pump. Interviews confirmed that outlying cases were from people who would regularly walk past the pump and take a drink. He removed the handle off the water pump and the outbreak ended almost overnight. The cause of cholera (bacteria Vibria cholerae) was unknown at the time, and Snow’s important work with cholera in London during the 1850s is considered to be the beginning of modern epidemiology. Some have even gone so far as to describe Snow’s Broad Street Map as the world’s first GIS. Although this is clearly an exaggeration, it does demonstrate the potential power that maps have when analyzing complex data.



    John Snow only had one cluster to identify and diagnose. Microsoft MapPoint can be used to speed Snow’s process of visual cluster analysis. Importing and plotting the data in MapPoint only takes minutes with MapPoint’s Data Import Wizard. Different pushpin colors can be used to plot different attribute values such as customer service ratings. By looking at the resulting map, you can quickly find potential problems, allowing you to find their root causes and fix them.

    Visual cluster analysis like this has its limitations. For example, it cannot perform a quantitative analysis, and it cannot be used to apply constraints on the final clusters. For example, you may wish to limit the spatial or numeric size of a cluster. Or you may wish to limit an attribute such as sales volume. This is where automated tools such as the MPCluster add-in for Microsoft MapPoint come in useful.

    A good application of geospatial cluster analysis is to analyze customer locations. If you discover that your customers are grouped in a particular location, you can investigate why. This information is then invaluable in your future marketing efforts. You can increase your marketing efforts, or identify future markets that have similar characteristics.

    MPCluster analysis can be combined with other MapPoint tools to produce more extensive analyses. Here is such an example that analyzes delivery catchment areas, courtesy of Lucassen Consulting:

    This map visualizes a supply chain (CA only) with delivery addresses, showing cluster analysis and subsequent drive time zones from the newly found clusters to show the catch areas. When you zoom in (see small map below), it is easy to identify the various key elements in the analysis. The underlying data are in Excel format.




    Detail of the San Francisco Bay Area:



    The second part of this article demonstrates the use of MPCluster to find clusters in another Lucassen Consulting dataset.

    Acknowledgments

    Thank you for Paul Lucassen of Lucassen Consulting for the above California maps.

    Further information on MPCluster including a free 14 day trial can be found at MPCluster: Cluster Analysis for Microsoft MapPoint
    Last edited by Eric Frost; 10-23-2013 at 10:14 AM.
    Winwaed Software Technology LLC
    http://www.winwaed.com
    See http://www.mapping-tools.com for MapPoint Tools

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Cluster analysis
    By pinem1 in forum MapPoint Desktop Discussion
    Replies: 6
    Last Post: 07-15-2010, 01:45 PM
  2. Replies: 0
    Last Post: 06-14-2010, 09:47 AM
  3. New Product Released: MPCluster Cluster Analysis for MapPoint
    By Winwaed in forum News and Announcements
    Replies: 0
    Last Post: 04-05-2010, 11:08 AM
  4. Virtual Earth Applications on Windows Azure (Part 1)
    By Johannes Kebeck's Blog in forum Bing Maps Blogs & Tweets
    Replies: 0
    Last Post: 02-27-2009, 09:50 AM
  5. The MapPoint Web Service Helper - Part I: Windows and Web Applications
    By Eric Frost in forum MP2K Magazine Articles
    Replies: 0
    Last Post: 01-05-2008, 09:43 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25