https://www.wikicshse.ru/index.php?action=history&feed=atom&title=Lecture_1._IntroductionLecture 1. Introduction - История изменений2026-06-06T12:13:10ZИстория изменений этой страницы в викиMediaWiki 1.45.3https://www.wikicshse.ru/index.php?title=Lecture_1._Introduction&diff=428&oldid=previmported>Katya: Migrated current public revision from wiki.cs.hse.ru2016-11-05T19:48:34Z<p>Migrated current public revision from wiki.cs.hse.ru</p>
<p><b>Новая страница</b></p><div>Ekaterina Chernyak, Dmitry Ilvovsky<br />
<br />
== Brief history of NLP ==<br />
<br />
* January 7, 1954 — Georgetown experiment. Russian to English machine translation;<br />
* 1957 — Noam Chomsky introduced "universal grammar";<br />
* since 1961 — Brown Corpus;<br />
* the late 1960's — ELIZA, a simulation of a psychotherapist;<br />
* 1975 — Vector Space Model by Salton;<br />
* up to the early 1980's — rule based approaches;<br />
* after the early 1980's — machine learning, corpus linguistics;<br />
* 1998 — Language Model by Ponte and Croft;<br />
* since 1999 — topic modeling (LSI, pLSI, LDA, etc);<br />
* 1999 — "Foundations of Statistical Natural Language Processing" by Manning and Shuetze;<br />
* 2009 — "Natural Language Processing with Python" by Bird, Klein, and Loper.<br />
<br />
== Major tasks of NLP ==<br />
<br />
* Machine Translation<br />
* Text classification<br />
** Sentiment analysis<br />
** Spam filtering<br />
** Classification by topic or by genre<br />
* Text clustering<br />
* Named entity recognition<br />
* Question answering<br />
* Automatic summarization<br />
* Natural language generation<br />
* Speech recognition<br />
* Spell checking<br />
* User study design and evaluation<br />
<br />
<br />
== NLP techniques ==<br />
<br />
* The level of characters:<br />
** Word segmentation<br />
** Sentence breaking<br />
* The level of words — morphology:<br />
** Part of speech (POS) tagging<br />
** Word sense disambiguation<br />
* The level of sentences — syntax:<br />
** Parsing<br />
* The level of senses — semantics:<br />
** Coreference resolution<br />
** Discourse analysis<br />
** Semantic role labeling<br />
** Synonymy detection<br />
<br />
== Main problems ==<br />
<br />
* Ambiguity<br />
** Lexical ambiguity:<br />
** Time flies like an arrow; fruit flies like a banana.<br />
* Syntactic ambiguity<br />
** Police help dog bite victim.<br />
** Wanted: a nurse for a baby about twenty years old.<br />
* Neologism: unfriend, retweet, instagram<br />
* Different spelling: NY, New York City, New-York<br />
* Non-standard language: HIIII, how are u? miss u SOOOO much:((((<br />
<br />
== About this course ==<br />
<br />
It covers the following topics:<br />
* Tokenization<br />
* POS tagging<br />
* Key word and phrase extraction<br />
* Parsing<br />
* Synonyms detection<br />
* Language sources<br />
* Topic modeling<br />
* Text visualisation<br />
<br />
You can try to use Python and R for various tasks.</div>imported>Katya