Rafael Reséndiz Ramirez
monday, April 27, 2015
Simple Word Prediction
Rafael Reséndiz Ramírez
Mon Apr 27 17:45:52 2015
Smoothing Models
The model can predict the words that continue in a sentence, showing what words have a higher probability of occurrence. In computational linguistic, an n-gram is a continuous sequence of n elements of a given sequence(text or voice). The n-grams are collected from a text or speech corpus.
Smoothing
It is necessary a good estimate for the probability space n-gram model.
Implementation Kneser-Ney and Stupid Back-off
The Kneser-Ney model were slow to display predicted words and accelerate the model with less code and relative loss of prediction accuracy. If you own a large corpus, the performance of the Stupid Backoff app displays a similar prediction accuracy Kneser-Ney with Smoothing.
Here mi databases, records and variables.
1. Körner, M. C. (n.d.). "Implementation of Modified Kneser-Ney Smoothing on Top of Generalized Language Models for Next Word Prediction Bachelorarbeit"", (September 2013).
3. J. Schalkwyk, D. Beeferman, F. Beaufays, B. Byrne,C. Chelba, M. Cohen, M. Kamvar, and B. Strope, "Your word is my command”: Google search by voice: A case study", in Advances in Speech Recognition, Amy Neustein, Ed., pp. 61–90. Springer US, 2010.
4. X. Lei, A. Senior, A. Gruenstein, and J. Sorensen, "Accurate and compact large vocabulary speech recognition on mobile devices", in Proceedings of Annual Conference of the International Speech Communication Association (Interspeech), 2013, pp. 662–665.
6. [Coursera Discussion Board](https://class.coursera.org/dsscapstone-001/forum) .