Machine Learning for Natural Language Processing

Andreas Vlachos

Given at the ALTA summer school in Machine Learning for Digital English Language Teaching

Tutorial Overview

In this tutorial we will cover key machine learning concepts for natural language processing. In the first part we will discuss methods for classification and discuss how to use them for language-related tasks. We will begin with the perceptron algorithm for linear models and work our way towards models and features capturing higher level intuitions. In the second part we will look at the problem of assessing plausibility of a sentence, as it underpins many tasks in language assessment. We will look at n-gram language models and how they can be combined with elements of syntax, as well as some recent developments with neural netwrok architectures.


(NOTE: These are the html/reveal.js versions of the slides which should be fine for the browser (tested in Chrome). If you want PDFs, just add ?print-pdf to the link and print. For jupyter+python dynamic versions (with interactive content) you need to clone the github repo.)