This presentation introduces a Shiny-based Next Word Prediction application. The app allows users to enter a phrase and predicts the next most likely word using a simple frequency-based language model.

This report explores large text datasets from blogs, news articles, and Twitter to build the foundation for a predictive text input application. The analysis includes basic summary statistics, data cleaning and preprocessing, and exploratory analysis of word frequencies. Unigram, bigram, and trigram models are developed to predict the next word in a sequence, along with a backoff strategy to handle unseen word combinations. The report also discusses considerations related to model size, memory usage, and runtime efficiency, and outlines plans for deploying the model as a Shiny application.

This presentation is a reproducible pitch for a Shiny application that demonstrates how the mean affects a histogram of Galton’s child height data. The app allows users to interactively adjust the mean using a slider and observe changes in the histogram and mean squared error (MSE). The presentation was created using R Markdown and ioslides and includes embedded R code.