Simple Self-Learning Words Prediction

postalC
19th April 2015

Introduction

Around the world, people are spending an increasing amount of time on their mobile devices for email, social networking, banking and a whole range of other activities. But typing on mobile devices can be a serious pain.

In this project, we work with our corporate partner Swiftkey, to builds a smart predict text model algorithm and software that would increase the user experince when type on their mobile devices.

The idea is simple, When someone types: “Go on a romantic date at the”. Then the software will presents at least 10 options for what the next word might be. For example the first three words might be:

  • Beach
  • Movie
  • Restaurant

Prediction Algorithm

  • Markov Chain
    A mathematical system that undergoes transitions from one state to another on a state space. Markov Chain
  • Katz's back-off model
    A generative ngram language model that estimates the conditional probability of a word given its history. Katz backoff model

Prediction Input :: How to use

  • Instruction Steps - (top left)
    project-info
  • Input your data - (2nd left)
    data-input
  • Searching for words - (botton right)
    search search-result

Prediction Resut :: How to read

  • Top 100 - next possible words - (bottom left/middle)
    top-10 top-11-100
  • Top 100 - Words Cloud - (top right)
    word-cloud