This project implements predictive text using Java, and it's part of the assignments of the Java programming principles of software design course. The code was originally developed in BlueJ, then moved to IntelliJ and adjusted to run there.
The fundamental ideal is to generate random text by training character selection on a training text, provided in the /data folder.
The code includes a set of Markov models of n orders, where n indicates the number of characters used to predict the following one during text generation.The code also implements an efficient version of the model, in which predicted characters are not computed from scratch at every run, but instead stored in a hashmap to be queried multiple times.