We are planning a new conference on computational modeling in linguistics, whose first meeting is tentatively scheduled to take place at UMass in Fall 2017. We have been hosting a discussion about various aspects of the plans for this conference on the UMass Computational Phonology Lab blog page and would love to get more input from linguists about our plans. We want to gather as much information as we can about how best to realize our goals for this conference and make it as accessible and useful to linguists as possible!

Please follow the link to our lab blog for full details and discussion, but to summarize, this conference is intended for linguists and cognitive scientists using computational and mathematical approaches to study the human language faculty.  We’re particularly interested in input from linguists on the following key aspects of the current plans:

• The short-term and long-term co-location/venue possibilities
• The name and ideal scope and target audience for the conference
• The tentative plan to have short paper (6-8pp) submissions (rather than abstract submissions). Please go to this post for discussion and comments on this aspect of the current plans.

We’ve had quite a bit of discussion on our blog already, primarily from potential participants of this conference who also regularly attend workshops and/or conferences affiliated with the ACL. This discussion has been very useful, but the conference would ideally be accessible and appealing to a broader community of linguists than is currently represented in the discussion. So we are looking for additional input from linguists who are potentially interested in this new conference, but who do not generally attend ACL – we’d be very grateful to hear about the kinds of considerations that could increase or decrease the chances phonologists and other linguists would attend and/or submit their best work to this conference.

Many thanks!
Gaja Jarosz

A CS+Linguistics group of collaborators — Su Lin Blodgett (CS), Lisa Green (Linguistics), and Brendan O’Connor (CS) — have published a paper on African-American English and Twitter. It will be presented at Empirical Methods in Natural Language Processing this November, and published in the Proceedings of EMNLP. A preprint is linked below.

Demographic Dialectal Variation in Social Media: A Case Study of African-American English

Abstract:
Though dialectal language is increasingly abundant on social media, few resources exist for developing NLP tools to handle such language. We conduct a case study of dialectal language in online conversational text by investigating African-American English (AAE) on Twitter. We propose a distantly supervised model to identify AAE-like language from demographics associated with geo-located messages, and we verify that this language follows well-known AAE linguistic phenomena. In addition, we analyze the quality of existing language identification and dependency parsing tools on AAE-like text, demonstrating that they perform poorly on such text compared to text associated with white speakers. We also provide an ensemble classifier for language identification which eliminates this disparity and release a new corpus of tweets containing AAE-like language.