search

euagendas / m3webdemo

A simple web demo of the m3inference library
http://www.euagendas.org/m3demo
GNU Affero General Public License v3.0
8 stars 1 forks source link

Extend description #16

Closed faflo closed 5 years ago

faflo commented 5 years ago

"M3 is a deep learning system that infers demographic attributes directly from social media profiles - no further data needed. This web demo showcases M3 on Twitter profiles, but M3 works on any similar profile data, in 32 languages. To learn more, please see our open-source Python library, m3inference or read our Web Conference 2019 paper for details."

computermacgyver commented 5 years ago

I have expanded the about text and tried to link to everyone's profiles online. Could everyone please check I have included the best link for them? I have also included a bit of text on the models.

          <h4>Learn more</h4>
            <p>
              M3 is a deep learning system that infers demographic attributes directly from social media profiles&mdash;no further data is needed. This web demo showcases M3 on Twitter profiles, 
              but M3 works on any similar profile data, in 32 languages. To learn more, please see our <a href="https://github.com/euagendas/m3inference">open-source Python library m3inference</a> 
              or <a href="https://doi.org/10.1145/3308558.3313684">read our Web Conference (WWW) 2019 paper</a> for details. The paper also includes fully interpretable multilevel regression methods that estimate inclusion probabilities uisng the inferred demographic attributes
              to correct for sampling biases on social media platforms.

            </p>

            <p>This web demo was created by <a href="http://www.oii.ox.ac.uk/people/hale/">Scott Hale</a> and <a href="https://www.oii.ox.ac.uk/people/graham-mcneill/">Graham McNeill</a>. 
            The plots are produced with the <a href="https://github.com/gjmcn/vizsla">Visla JavaScript API for Vega-Lite</a>, and the <a href="https://github.com/euagendas/m3webdemo">source code</a> for this webapp is freely available.
            The research leading to M3 was carried out by 
            <a href="https://github.com/zijwang">Zijian Wang</a>, 
            <a href="http://www.oii.ox.ac.uk/people/hale/">Scott A. Hale</a>, 
            <a href="https://www.lsv.uni-saarland.de/?id=231">David Adelani</a>, 
            <a href="https://www.cics.umass.edu/people/grabowicz-przemyslaw">Przemyslaw A. Grabowicz</a>, 
            <a href="https://www.researchgate.net/profile/Timo_Hartmann">Timo Hartmann</a>, 
            <a href="https://www.gesis.org/person/fabian.floeck">Fabian Fl&ouml;ck</a>, and 
            <a href="https://www.si.umich.edu/people/david-jurgens">David Jurgens</a>.
          </p>
computermacgyver commented 5 years ago 
       <a href="https://www.lsv.uni-saarland.de/?id=231">David Adelani</a>,