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

Mining the inbox: Building an email priority algorithm

  • Mar 6, 2014 · 6:30 PM
  • This location is shown only to members

How does Gmail decide which of your emails are important? Can a computer determine whether an email contains links to cat pictures or an urgent request from your boss? With modern data mining techniques, we can train a model to rate an email’s priority/importance. In this talk, we describe the major steps and data mining algorithms involved in designing and implementing a priority inbox. These include assembling the training data, tuning the model, and deploying to a production environment. Some basic knowledge about machine learning and python will be helpful - code examples will be available on Github.

About the presenter:

Max Mautner is a Senior Analyst at AdHarmonics, a Cambridge firm that matches auto insurance providers with customers online. He focuses on email, search, and display marketing. He has worked at Cogo Labs (Cambridge) on email marketing, as a derivatives trader in Chicago, and as a mathematics researcher at the National Science Foundation. He has a BA in Mathematics from Claremont McKenna College

Join or login to comment.

  • Bobby M.

    Any chance video or slides can be shared?

    March 7, 2014

  • vitaly b.

    Nice presentation. Just want to mention one thing. A drawback of using a third party (Google in this case) labeled corpora will most probably not allow you to reach a decent accuracy unless you use their features in classification.

    March 7, 2014

  • Demetri S.

    Great talk!

    March 6, 2014

  • A former member
    A former member

    Hi hi

    March 6, 2014

  • Barbara

    Hello, looking forward to meeting everyone

    March 4, 2014

  • sands f.

    Apologies for the off-topic ping, but if anyone knows the guys and gals at Recorded Future, I met someone at the Machine Learning hackathon at Hack/Reduce and can't find contact information. Any help would be appreciated. Thanks!

    1 · February 24, 2014

  • A former member
    A former member

    Thanks for rescheduling this event. Looking forward to it.

    February 14, 2014

  • Sheamus M.

    We made the call to cancel and you should see the rescheduling notification email for March 6th. Thanks to Max and Adrienne at hack/reduce for being accommodating in rescheduling this event.

    February 13, 2014

  • R.J. P.

    Hello, would it be possible to get a confirmation that this Meetup is a go? Inquiring minds want to know.

    February 13, 2014

    • A former member
      A former member

      Last I heard it's in the hands of the venue to make the call. I'm sure Sheamus or Daniel will send out an email shortly about whether the talk will be rescheduling or not

      February 13, 2014

    • Oliver R.

      By 6:30pm snow will have turned into rain.

      February 13, 2014

  • A former member
    A former member

    This is interesting indeed, but big snow storm tomorrow. I hope there's a reschedule for this. :-)

    February 12, 2014

  • David P.

    Hi.. this meetup sounds very interesting

    February 7, 2014

Our Sponsors

  • Strata + Hadoop World

    20% special discount code UGBOSDM

  • WeWork

    Contact WeWork, mention your BDM Meetup Membership for special discounts

  • Open Data Science

    Signup for ODSC Newsletter. Receive attractive ODSC East Conf. offers!

People in this
Meetup are also in:

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