After taking off the month of July, DC NLP is back with a bang in August! We'll be meeting at Comcast Labs DC again this month with two fantastic presentations on Deep Learning for NLP and advanced Search Engines. Also, courtesy of our friends at O'Reilly Media, we'll have a bunch of swag to give away including books, t-shirts, and other assorted goodies. This is sure to be an event you won't want to miss!
First up, Ari Chanen (https://www.linkedin.com/in/ari-chanen-phd-3390253), Lead AI Engineer at MITRE (https://twitter.com/MITREcorp), will give a talk about his recently-published paper entitled "Deep Learning for Extracting Word-level Meaning from Safety Report Narratives" which he presented at the aviation-focused 2016 Integrated Communication Navigation and Surveillance (http://i-cns.org/) conference. This work is focused on practical questions surrounding the use of the deep learning technique word2vec. The first novel technique Ari will discuss is a way of picking a reasonable number of dimensions for a word2vec model which is based on the level of disagreement between a set of models trained on the same data. No training data is needed for this technique. The second novel technique is using multiple word2vec models together in an ensemble model to get a more robust set of similar terms over using a single word2vec model.
For our second talk, Mark Lidd (https://www.linkedin.com/in/kramdl), Team Lead at DataFission (https://twitter.com/datafission), will give a presentation on DUSE: a content-based search engine for digesting and searching unstructured data (http://www10.giscafe.com/blogs/gissusan/2016/07/29/datafissions-new-duse-search-engine-lets-you-search-any-unstructured-data/). DUSE allows users to search anything digitally, from video, audio, network traffic, satellite images, radar data, malware, and images -- as well as unstructured text -- all in one search. Unlike traditional search engines such as Yahoo, Google, and Microsoft that have been around for 15 years, searching with DUSE doesn't require further work to search the medium the user requires. It contains both simple APIs and GUIs that provide rankings and answers to complex data queries to uniforms and analysts, as well as advanced controls for data scientists, such as access to live data structures and tools at a very deep level.
A brief note from O'Reilly Media:
Discover the real-world opportunities of applied artificial intelligence
AI is (finally) driving an explosion in intelligent software—bots, agents, voice and IoT interfaces. Learn how to implement AI in real-world projects today and explore what the future holds for intelligence engineering. Save 20% with discount code UGDCNLP. Check out the impressive agenda and speaker lineup. http://www.oreilly.com/pub/cpc/21357
We'll gather starting at 6:30pm at Comcast Labs (http://dclabs.comcast.com/) - someone will greet you outside and let you into the building. Introductions & announcements will start around 7:00pm, and presentations will begin at 7:30pm. Afterwards, there will be plenty of time for follow-up questions, networking, and more.
DC NLP meets each month to network, socialize, and learn about the interesting work folks are doing in natural language processing, computational linguistics, text analytics, and more.
Do you have something you'd like to share with the group? Let us know! We're always looking for speakers to give talks at future meetups, and don't forget to follow @DCNLP (https://twitter.com/DCNLP/) on Twitter!