Marcelo Tibau
March 10, 2017
Based on Google's stupid backoff algorithm.
Despite its name, it performs almost as well as some more complex models given very large data. On the authors' own words: Stupid Backoff is inexpensive to calculate in a distributed environment while approaching the quality of Kneser-Ney smoothing for large amounts of data.
For more details: Brants, Thorsten; Popat, Ashok C; Xu, Peng; Och, Franz J; Dean, Jeffrey. “Large language models in machine translation”. EMNLP/CoNLL. 2007.
An interactive web app that takes in text input and return the predicted upcoming terms.
Step right up at: https://marcelotibau.shinyapps.io/word_predictor/