StatNLP

$\def\prob{p} \def\vocab{V} \def\params{\boldsymbol{\theta}} \def\argmax{\mathop{\arg\,\max}} \def\param{\theta} \def\bpi{\boldsymbol{\pi}} \def\balpha{\boldsymbol{\alpha}} \def\bbeta{\boldsymbol{\beta}} \def\perplexity{PP} \def\x{\mathbf{x}} \def\y{\mathbf{y}} \def\Xs{\mathcal{X}} \def\Ys{\mathcal{Y}} \def\perplexity{PP} \def\train{\mathcal{D}_\mathit{train}} \def\counts#1#2{\#_{#1}(#2)} \def\RR{\bf R} \def\length#1{\mathop{length}(#1)} \def\aligns{\mathbf{a}} \def\align{a} \def\source{\mathbf{s}} \def\ssource{s} \def\target{\mathbf{t}} \def\starget{t}$

Introduction

What is NLP?

Building computer systems that analyze, understand, and generate natural languages.
Deep understanding of broad language
- not just string processing or keyword matching

Exercise

Can you think of NLP applications?

Speech Recognition

Speech Recognition

Speech Recognition is usually not considered NLP. We will not cover this topic here.

Machine Translation

Personal Assistants

Information Extraction

Information Extraction

Summarization

Question Answering

Sentiment Analysis

Machine Comprehension

Cognitive Science and Psycholinguistics

Why is it difficult?

Ambiguity Everywhere

0

Fed raises interest rates 0.5% in effort to control inflation

Fool a Sentiment Analyzer

Fool a Machine Translator

Count N-grams

Syllabus

This course covers three core dimensions of NLP:

Tasks (tokenization, language models, machine translations etc.)
Methods (Maximum Likelihood, EM algorithm, dynamic programming etc.)
Implementations (document representations, efficient feature extraction etc.)

The lectures, and the structure of the course, will be driven by tasks.

NLP Tasks

Tokenization, Segmentation
Language Modelling
Machine Translation
Syntactic Parsing
Document Classification
Information Extraction
Topic Models and Word Embeddings

Questions?