Lexalytics®, the leader in “words-first” machine learning and artificial intelligence, and the University of Massachusetts Amherst announced today that they will present their paper, “Efficient Graph-based Word Sense Induction by Distributional Inclusion Vector Embeddings,” at TextGraphs-2018, a workshop series at The 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) on June 6 in New Orleans.
Who: UMass Amherst graduate students Haw-Shiuan Chang, Amol Agrawal, Ananya Ganesh, Anirudha Desai and Vinayak Mathur; Andrew McCallum, Professor and Director of the Center for Data Science at UMass Amherst; and Alfred Hough, Lead AI Researcher at Lexalytics
What: Workshop on “Efficient Graph-based Word Sense Induction by Distributional Inclusion Vector Embeddings”
When: Wednesday, June 6, 2018, 4:00 p.m. ET — 4:25 p.m. ET
Where: NAACL HLT, New Orleans, Louisiana (Hyatt Regency New Orleans, Bolden 1)
Why: The paper describes a set of algorithms for Natural Language Processing (NLP) that match or exceed the state of the art on several evaluation tasks, while also being much more computationally efficient at doing the following:
- From a large unannotated text corpus, discover the multiple meanings of polysemous words (e.g., bank, the financial institute vs. bank, the side of a river)
- Assign each meaning, or “sense,” a “word embedding,” a numeric representation of the meaning of the word
- Provide related words for each sense to make it clear what the meanings are
- Identify the correct sense of a polysemous word in new sentences
This new approach to word-sense induction comes from the work of the Lexalytics Magic Machines AI Labs, launched in 2017 in partnership with the University of Massachusetts Amherst’s Center for Data Science and Northwestern University’s Medill School of Journalism, Media and Integrated Marketing Communications to drive innovation in AI.
About UMass Amherst
UMass Amherst, the Commonwealth’s flagship campus, is a nationally ranked public research university offering a full range of undergraduate, graduate and professional degrees. The University’s College of Information and Computer Sciences (CICS) is nationally recognized for its research activities, and has one of the highest ranked and most competitive graduate research programs in the nation. The Center for Data Science, housed within CICS, is a leading interdisciplinary hub for research, education and industry collaboration in data science. At the Center’s launch in April 2015, the University announced a projected investment of $100M, with plans to hire 40 additional faculty over the coming decade to augment the over 150 data science-related faculty in the Five Colleges. The Center’s Industry Affiliates Program includes many actively engaged companies such as Amazon, Google, IBM, MassMutual, Microsoft, Oracle, and Pratt & Whitney, as well as a growing number of local startups. The Center is creating new research initiatives bringing together faculty research, industry partners, entrepreneurs, philanthropies, and state/federal government in areas including healthcare, education, workforce analytics, energy, agriculture, conservation, machine learning, information integration, scalable systems, computer vision, and human language technology.