KDnuggets Home » News » 2014 » Apr » Publications » Where are your users? Geo-localization with KNIME ( 14:n10 )

Where are your users? Geo-localization with KNIME


Learn how KNIME can help you improve user understanding through Geo-localization of IP addresses and dynamic visualization. Access free white paper for more details.



By Rosaria Silipo, Apr 28, 2014.

Planning Community Events around the World
It is vital for a company to know where its user base is located. User geo-localization can be easily implemented using the Open Street Map extension and its results can be easily visualize, either as a static map or as a movie, using the Image Processing extension of KNIME. Both extensions have been made available by the KNIME community. The goal of this study is to get more insight about the KNIME user base geographical distribution, in order to better plan for future community events.

The Apache Weblog file
Such geographical information is available via the connecting IP addresses. From the Apache weblog file of the KNIME web server, visitor data, including their IP addresses, are extracted. The time window is restricted to the week following a new KNIME release. In fact, during these days, regular KNIME users are more likely to download the latest version of the tool, producing a more reliable picture of their geographical distribution.

Geo-localization of IP addresses
IP addresses are mapped to geographical locations and displayed on a world map using the KNIME Open Street Map integration extension, made available by the KNIME community. The world map can either host a summary of all KNIME downloads during the selected week or it can be re-drawn for each day producing a world map sequence. Geo-location of KNIME downloads
Dynamic visualization of the daily downloads around the world
The visualization of a long sequence of map images can be problematic on a static report, but not in a movie. A particular node in the KNIME Image Processing extension, also made available by the KNIME community, converts image sequences to movies. This node is used and a movie is produced, showing the dynamic evolution of KNIME downloads around the world and day by day following a new release. knime-geolocalization-movie
 
Whitepaper, Workflows, and the KNIME Tool
This very interesting study covers weblog reading, geo-localization, and image processing to represent the dynamic distribution of the KNIME downloads in time and around the world. All workflows are available on the KNIME EXAMPLES public server under “008_WebAnalytics_and_OpenStreetMap”; the whitepaper can be found at knime.org/files/geolocalization_of_knime_
downloads.pdf
; and the KNIME software can be downloaded for free from knime.org

Dr. Rosaria Silipo (Linkedin) is not only an expert in data mining, machine learning, reporting, and data warehousing, she has become a recognized expert of the KNIME data mining engine, on which she published three books: ( KNIME Beginner's Luck, The KNIME Cookbook, The KNIME Booklet for SAS Users).
Previously Dr. Silipo worked as a freelancer data analyst for many companies in Europe. She led the SAS development group at Viseca (Zürich, Switzerland), implemented the speech to text and the text to speech interfaces at Spoken Translation (Berkeley, USA), and developed a number of speech recognition engines in different languages at Nuance Communications (Menlo Park, USA). She got her doctorate in biomedical engineering in 1996 from U. of Florence (Italy).

Sign Up