Introduction to Python and Natural Language Technologies

Lecture 01 - Introduction

September 6, 2017

1.1 About this course

1.2 Requirements, course material

1.3 What is Natural Language Processing?

1.1 About this course

Lecture: Wednesday, 12.15-14.00, Rm QBF14

Lab: Friday, 12.15-14.00, TBA

Weeks 1-6: "Advanced Introduction" to Python

Weeks 7-14: Introduction to Natural Language Processing (NLP)

Detailed schedule available at https://github.com/bmeaut/python_nlp_2017_fall

Instructors (1)

Judit Ács (BME AUT), judit@aut.bme.hu

Gábor Borbély (BME Dept. of Algebra), borbely@math.bme.hu

Instructors (2)

Dávid Nemeskey (MTA SZTAKI), nemeskeyd@gmail.com

Gábor Recski (BME AUT), recski@aut.bme.hu

General questions, e.g. about requirements, grading, etc. should be sent to Gábor Recski (recski@aut.bme.hu, +36209709419)

1.2 Requirements, course material

Requirements

3 homework assignments, each of which must be awarded a passing grade

Late submission of 1 homework is possible by the end of the re-take period (pótlási hét).

Oral exam (NLP + Python, also in English!)

Final grade: 50% homeworks, 50% exam

See also the official course description

Course Material

All material used in class is available from the course's GitHub repository at https://github.com/bmeaut/python_nlp_2017_fall

GitHub issues should be used to ask questions and discuss topics related to the course

Links to external material will be spread out across the slides

Official Python documentation at https://docs.python.org/3/

Jurafsky, D., & Martin, J. H. (2014). Speech and language processing. Pearson
- 2nd edition
- 3rd edition (work-in-progress)

1.3 What is Natural Language Processing?

What is NLP? (1)



In [2]:

    
from graphviz import Digraph
dot = Digraph()
dot.graph_attr['rankdir'] = 'LR'
dot.node('I', 'natural language input')
dot.node('C', 'computer', shape="square")
dot.node('O', 'arbitrary output')
dot.edge('I', 'C')
dot.edge('C', 'O')

dot.node('I2', 'arbitrary input')
dot.node('C2', 'computer', shape="square")
dot.node('O2', 'natural language output')
dot.edge('I2', 'C2')
dot.edge('C2', 'O2')

dot.render("nlp")
dot









    Out[2]:

What is NLP? (2)

	less abstract	middle	abstract
language:	Natural Language	Artificial Language	Mathematics
example:	English	programming langauges	formulas, equations
usage:	spoken by many	fewer	spoken by few
interpretation:	vague	mostly well defined	only one way to interpret
applications:	everywhere, by everyone	broad, but the users don't see it	specific, unseen by users

What is NLP? (3)

"NLP is a branch of artificial intelligence" (LifeWire)

"Natural language processing (NLP) is a field of computer science, artificial intelligence and computational linguistics concerned with the interactions between computers and human (natural) languages" (Wikipedia)

"[NLP] is one of the most important technologies of the information age. (...) Applications of NLP are everywhere because people communicate most everything in language" (Manning & Socher: NLP with Deep Learning course at Stanford)

"Ever wondered how Google Translate works, or how companies do automated resume processing? Want to build a computer that understands language? This course is for you. It develops an in-depth understanding of both algorithms for processing linguistic information and the underlying computational properties of natural languages." (Manning et al.: Intro to NLP course at Stanford)