addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscontroller-playcrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1light-bulblinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonprintShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Unsupervised Neural Networks through Self-Organizing Maps in R

Dear Toronto R user group,

I'm very pleased to announce the next Toronto R user’s group meeting. Our guest speaker, Masoud Charkhabi, will be giving a talk about Unsupervised Neural Networks through Self-Organizing Maps in R, with interesting Applications in Text Mining and Data Visualization.

Masoud Charkhabi is Senior Manager of Advanced Analytics at the Canadian Imperial Bank of Commerce. Masoud leads a group of Data Science practitioners that specialize in deriving insight from massive data sources using Statistical Learning techniques. An important challenge in industry is integrating unstructured and structured data sources, then being able to articulate findings from analysis on the high-dimensional datasets that result from this aggregation. This presentation will attempt share applications of R packages to apply the Kohonen SOM technique to these problems.

Venue kindly provided by my fellow co-organizer, Professor Murtaza Haider, at the Ted Rogers School of Management at Ryerson University.

The capacity for the room is about 60, so we’ll have to prioritize in order of sign-up for the meeting. There is no charge for this meeting.

When: Friday December 7, 2012 at 3:00 - 5:00 pm

Where:Ted Rogers School of Management at Ryerson University
55 Dundas Street West
Lecture Hall, 9th floor, Room TRS 3-129
Toronto, ON M5G 2H1

Please RSVP to this meeting here:

This is a great opportunity not to be missed. Looking forward to seeing you there!

Leo Guelman.

Join or login to comment.

  • Hardik

    Please post the presentation link! Thank you all

    December 10, 2012

  • Mujahid S.

    This was an excellent presentation on the topic. I have been very active user of SOM's for over a decade and really appreciate Masoud for putting this all together in this short presentation.

    We have used SOMs for Microarray data analysis, below is link to my publication
    The algorithm is called Binary Tree-Structured Vector Quantization (BTSVQ)
    The Matlab code is freely available, please contact me and I will be happy to provide.

    BTSVQ goes one step further to cluster the similar neurons automatically. If the data is very high dimensional or if the feature space is large, then visually clustering neurons is quite a challenge. (Which we were facing for the Microarray data - the features are patients)

    December 10, 2012

  • Gustavo A.

    Interesting presentation. Thanks to Masoud and the organizers.

    December 8, 2012

  • Alexey

    Thanks a lot for the great presentation!
    Is there any way the powerpoint slides could be made available for downloading? I would like to read some of those references.
    Also at the end someone commented that they successfully used SOMs for cancer diagnosis. Any chance you could provide a reference for that? I would really like to read that paper.
    Thanks a lot!

    1 · December 8, 2012

  • Adam M.

    A really well structured presentation. Thanks!

    December 7, 2012

  • Monica W.


    December 7, 2012

  • Hardik

    I am sorry I wont be able to make it. I have a humble request to the group can you please arrange session later in the day from 5 to 7 , instead of 3 to 5, people like me who want to come, but can't due to full-time job

    December 6, 2012

  • Adam M.

    I'm sorry I missed the last one. I am really looking forward to this one. It will be nice to exercise the R instead of the Apache Mahout skills for a change.

    November 7, 2012

Our Sponsors

  • R Consortium


Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy