This repository holds scripts for creating a language model and a Shiny web application for a simple word predictor. It was created to complete the Coursera Data Science Specialization Capston project and is my first R project and Shiny App.
You can try the app at: https://codrin.shinyapps.io/WordPredictor/
- Data Science Capstone Notebook Text Prediction App Quanteda DT.Rmd holds code for language model creation.
- TODO holds some code for calculating and plotting some basic text characteristics.
- TODO holds code for a wordPredictor function that is shared between language model script and app.
- Word Predictor folder contains files necessary for the Shiny App
- Data Science Capstone Text Prediction.Rpres is a short presentation about the language model and app development.
- Other .Rmd notebooks are older versions of the language model script; for some reason my Rmd keep breaking and cannot knit when I break off processes. Still have to figure out what is causing this to find a solution.
Still lots to be fixed and optimized:
- Recheck language model creation steps
- Check profanity filter
- Reduce repetitive code in data_descriptives.R and wordPredictionFunction.R
- Do elaborate code profiling to improve computational efficiency