Effective Word Prediction in Urdu Language Using Stochastic Model

Authors

  • M. Farhan Siddiqui Department of Computer Science, University of Karachi, Pakistan
  • M. Hassan Department of Computer Science, University of Karachi, Pakistan

DOI:

https://doi.org/10.30537/sjcms.v2i2.304

Keywords:

Word Prediction, Natural language processing, stochastic model, unigram, interpolation, markov model.

Abstract

Word prediction and word suggestion is an important tools for writing contents in any language, in which the right word is predicted in a current context. Writing contents on English keyboards to produce other language contents is too hard and time consuming for everyone and requires more practice. To increase the typing speed especially in mobile and smart phones or for creating contents on social networks derived the need of this tool in every language. This paper presents a state of the art research for word prediction in Urdu Language (UL) based on stochastic model.  Hidden Markov Model was implemented to predict the next state, while Unigram Model was also used to suggest the current state and  the next hidden state,  N-Gram Model was followed keeping N=2. The tool is developed to implement this model for Urdu Language (UL) and tested by regular and new URDU content writers to check their improvements in their typing speeds.

Downloads

Download data is not yet available.

Downloads

Published

2019-03-09