This project presents a Shiny web application that predicts the next word in a given English phrase using an N-gram language model. The app takes user input and suggests the most likely next word based on patterns learned from a large corpus of text (e.g., Twitter, blogs, and news data). The underlying algorithm leverages quadgrams, trigrams, and bigrams with a backoff strategy to ensure accurate and fast predictions. This tool demonstrates a simple yet effective approach to predictive text modeling, similar to autocomplete features used in modern keyboards.

This report presents an exploratory analysis of three English text datasets (blogs, news, and Twitter) for a word prediction project. It includes basic statistics, data cleaning, visualizations of common word usage, and outlines the plan for building an n-gram-based predictive text model. The final model will be deployed in an interactive Shiny app to predict the next word based on user input. The report is written for a general audience and highlights key findings to support algorithm development.