Issue 1 - NLP Legend dies, Facebook’s wav2letter, Tensorflow for Deep Learning Research …
Released Date: January 12, 2018
Aravind Joshi, revered by many as a legend in NLP for his work on understanding how language is represented by our minds and how computers should process and model language, passed away before the new year. The entire AI and NLP community mourns the lost of a true legend.
Source: Language Log
On Code and Data...
Earlier last week, Facebook AI research team released wav2letter — “a simple and efficient end-to-end Automatic Speech Recognition (ASR) system.” The toolkit provides pre-trained models that can get you started right away with transcribing speech. The accompanying paper can also be found here.
Analysis conducted on NLP/ML conferences. Include charts containing information on trending topics and conferences.
Source: Marek Rei Blog
Open dataset that includes patent grants released by Gensim.
On Education and Research...
New course, “Tensorflow for Deep Learning Research”, opened by Stanford.
An new course on Natural Language Processing with NLTK for beginners.
Source: NLP for Hackers
This paper dicusses how to obtain state-of-the-art results on a sentiment analysis task with simple Naive Bayes and Support Vector Machine variants.
Source: Baselines and Bigrams: Simple, Good Sentiment and Topic Classification
3rd edition of “An introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition” by Daniel Jurafsky (Stanford University) and James H. Martin (University of Colorado at Boulder).
A new book entitled “Introduction to Linguistics for Natural Language Processing.”
This paper discusses a new approach to sarcasm detection using a multi-faceted approach: eye-tracking features, NLP, and deep learning algorithms. This is part of a seminar series offered at Standford.
Source: Sarcasm Detection: A Computational Cognitive Approach
Mercedes-Benz release MBUX, said to be a “revolutionary” and “best conversation you’ve ever had with a car” experience, all made possible by advanced natural language understanding and AI.
Source: Slash Gear
If you spot any errors or inaccuracies in this newsletter please open an issue.
Submit a pull request if you would like to add important NLP news here.