Hashtags and HR: The 7th NLP Dublin Meetup

After a brief break, we are very excited to announce our seventh NLP Dublin meetup. We are proud to go back to our beginnings and be hosted by our first ever host Nitro (https://www.gonitro.com/) at their EMEA HQ in Burlington Road.

https://cdn.gonitro.com/images/sites/nitroweb/pages/misc/nitro-brand-2015/nitro_dark.png

We have two great talks lined up by two great speakers from both academia and industry. On the academic side, Georgiana Ifrim will talk about Hashtagger+, an efficient learning-to-rank framework that allows the real-time tagging of news streams. On the industry side, Adrian Mihai will talk about how natural language understanding can be leveraged to disrupt HR technology. We are looking forward to seeing you all at the event!

AGENDA:

[18:00 - 18:30] Registration and networking

[18:30 - 18:35] Welcome message

[18:35 - 19:15] Dr. Georgiana Ifrim (https://www.cs.ucd.ie/AcademicProfile/GeorgianaIfrim/) (Assistant Professor, UCD): Hashtagger+: Efficient High-Coverage Social Tagging of Streaming News

News and social media now play a synergistic role, and neither domain can be grasped in isolation. On one hand, platforms such as Twitter have taken a central role in the dissemination and consumption of news. On the other hand, news editors rely on social media for following their audience's attention and for crowd-sourcing news stories. In this work we propose Hashtagger+, an efficient learning-to-rank framework for merging news and social streams in real-time, by recommending Twitter hashtags to news articles. We analyse a variety of machine learning models and show that, for the task of streaming hashtag recommendation, pointwise learning-to-rank is more effective than topic modeling and other more complex learning-to-rank approaches. We present new techniques for bootstrapping our learning-to-rank approach to address cold-start in recommendation for new articles and show that by using past recommendations we can improve the efficiency and coverage of our model.
In our comprehensive evaluation on real-data we show that we drastically outperform the accuracy and efficiency of existing methods. Our prototype system delivers recommendations in under 1 minute, with a Precision@1 of 94% and article coverage of 80%. By effectively linking the news stream to the social stream, via the recommended hashtags, we open the door to solving many challenging problems related to story detection and tracking.

[19:15 - 19:55] Adrian Mihai (https://opening.io/about-us) (Co-founder & CTO, Opening.io): Natural language understanding in HR technology

Adrian will give us a walk-through of an end-to-end resume analysis pipeline. He will talk about its different components: parsing, similarity matching and no-frills deep learning models deployment at scale.

[19:55 - 21:00] Networking and Pizza

IMPORTANT NOTES:

• If you can't make it, please RSVP to "NO" as soon as possible so that other people can take your place.

• If you are interested in speaking or sponsoring this event, please contact the event organizer.

Co-organized by Aylien (http://aylien.com/)

https://secure.meetupstatic.com/photos/event/8/5/c/1/600_451654241.jpeg