A simple NLP-based predictive text model using R and Shiny.

A predictive text model built using NLP and Shiny in R. This app takes an input phrase and predicts the next word using a simple n-gram language model. The presentation explains the algorithm, model logic, app design, and possible future improvements.

This project performs an exploratory data analysis (EDA) on the SwiftKey dataset, which contains text samples from Twitter, blogs, and news sources. The analysis involves loading and cleaning raw text data, generating descriptive statistics, and visualizing word usage patterns across datasets. Through tokenization, stopword removal, and frequency analysis, the report highlights linguistic trends that can inform the development of a next-word prediction model and Shiny app. The study demonstrates practical text mining techniques in R, including data sampling, regex cleaning, and visualization using ggplot2 and tidytext.

A predictive text model using NLP and Shiny, demonstrating real-time next-word prediction deployed as a web application.