sweitzen / yap-tm

Yet Another Predictive Text Model (YAP™) - A Shiny text prediction app

Yet Another Predictive Text Model (YAP™)

===========================================

YAP™ was implemented as the Data Science Capstone project for the Johns Hopkins University Coursera Data Science Specialization.

The YAP™ app uses a 5-gram language model to predict the next word from user input, ranking the predictions using a method called Stupid Backoff, as described in 'Large Language Models in Machine Translation' by T. Brants et al, in EMNLP/CoNLL 2007.

The data supporting this project can be found at

• https://d396qusza40orc.cloudfront.net/dsscapstone/dataset/Coursera-SwiftKey.zip

---

The project milestone report is available at:

• http://rpubs.com/sweitzen/data-science-milestone

The final product pitch is available at:

• http://rpubs.com/sweitzen/yaptm_pitch

A live demo of YAP™ is available at shinyapps.io:

• https://sweitzen.shinyapps.io/yap-tm/

---

Docker Compose

If you wish to run locally in a Dockerized Shiny Server:

version: '3'

services:
  yaptm:
    image: sdweitzen/yap-tm
    ports:
      - '3838:3838' 
    volumes:
      - shiny-db:/srv/shiny-server

volumes:
  shiny-db:

Once the Docker app is up, YAP™ will be available in your browser at localhost:3838.