Natural Language Learning (Strojové učení a přirozený jazyk)

Evaluation<50 F, <60 E, <70 D, <80 C, <90 B, >=90 A

Keyness (keywords and beyond), essays - Oct 11 10 points
Questionairre (Basics of machine learning) - 10 points
A chapter from Keyness in Texts (M. Bondi & M. Scott, eds.) - max. 10 points
Project - max. 45 points
Final exam - max. 30 points
Project report written in English - max. 10 points

Subjects at LF MU : Martin Komenda, Petr Dluhoš
Theses at FI MU : Šimon Suchomel, Andrej Verbó
Practice - TUL : Nikolay Dvinskikh, Andrej Herich
Shakespeare plays : Jakub Čech, Michal Petrov, Libor Vanek

Slides (here)

Osnova

• Uvod. Korpusy. Nástroje ppt, pdf
• Jazyk Prolog pro reprezentaci dat a znalosti
  Uvod do logickeho programovani pdf, ps, ps: 4 strany na A4
  YAP Prolog
  Metainterprety
  Gramatiky definitnich klauzuli
• Přehled metod
  Úvod do strojového učení ppt, ps, ps: 6 stran na A4i
  Zdroj: T. Mitchell, Machine Learning (slidy,SW)
  Version spaces (animation)
  Decision tree learning (animation 1) (animation 2)
  Induktivní logické programování ppt, systém Aleph
  Skryté Markovovy modely(Univ. of Leeds)
• Desambiguace
  Morfologická desambiguace(pdf)
  Recognition and Tagging of Compound Verb Groups in Czech
  PoS Tagging, ILP and Active Learning
  POS The State of the Art
  Penn Treebank tags

  *On Disambiguation in Czech Corpora(see just for fun)
  *Desambiguace pádů v baskičtině
  Desambiguace významu slov

• Kategorizace dokumentů
  Definice úlohy, předzpracování, učicí algoritmy, filtrace dokumentů
  Categorization of Czech Documents
  Výběr termů, metriky
  Konstrukce nových atributů
  Fragments and Text Categorization
• Sentiment analysis
  Christopher Potts' course on Sentiment lexicons
  Extracting Social Meaning and Sentiment(Stanford)
  Sentiment Analysis Symposium
  Lillian Lee Cornell Univ.
  Bo Pang Yahoo
• Extrakce informace z textu
  Definice úlohy
  Extrakce informace z odborných textů
  Interakce genu
  Genic Interaction Extraction Challenge
  Proceedings of LLL'05 Learning Language in Logic
  Xemic: Systém pro podporu rozhodování v analytické chemii, DP Michal Očko
  Identifikace situací a akcí v novinových zprávách
  
• Keyness
  Mike Scott page
• Text mining

• PAN Uncovering Plagiarism, Authorship, and Social Software Misuse
• Dotazovací systémy (Question-Answering systems) Communication with WWW in Czech
• Dolování z hypertextu
  WWW: Dolování z obsahu a z logů
  Klasifikace XML dokumentu
• Sémantický web
  Semantic Web Mining (Arizona State University)
  Some Ongoing KBS/Ontology Projects and Groups
  Deborah L. McGuinness, Stanford

  A Semantic Web Primer by Grigoris Antoniou and Frank van Harmelen, MIT Press 20004
  Web Mining: From Web to Semantic Web (LNAI 3209, Springer)
  The Semantic Web: Research and Applications (LNCS 3532, Springer)
  Information Sharing on the Semantic Web by Heiner Stuckenschmidt, Frank Van Harmelen, Springer

Literatura: ppt, ps, ps: 6 stran na A4

Teaching

On-line course on machine learning Stanford
On-line course on artificial intelligence Stanford

Work

Hunter: A Practical Guide to Critical Thinking: Deciding What to Do and Believe
Apprentissage automatique pour le TAL Special issue of TAL 2009, vol 50, No 3

People

Natural Language Processing by Ray Mooney
Ingo Feinerer TM package in R
Machine Learning and Natural Language Course by Dan Roth

Data sources

Weka archive containing 19 text datasets
Reuters21578
The 20 Newsgroups data set
NIST Text Analysis Conference Data

Toolkits

The Stanford Natural Language Processing Group Software
Online parsing/tagging web-applications for French, English and English and others
Universal Catalogue of tools and language sources
Freeling mostly for Spanish
Calais Information Extraction

British National Corpus
Czech National Corpus

LoogleFight vs. GoogleFoght

GATE
GATE-Weka
JWPL Java Wikipedia Library
Natural Language Toolkit
NLP Tool Suite by Katrin Tomanek, Jena
NooJ by Max Silberztein
OpenNLP

International Journal of Corpus Linguistics
DBpedia
StatSoft, Inc. (2010). Electronic Statistics Textbook
Symbolic learning of natural language by Dimitar Kazakov and James Cussens at ESSLLI'04
Constraint grammar
ILP Applications
Introduction to NLP, Jan Hajic
Natural Language Learning
National Centre for Text Mining. Manchester
Special Issue for the Journal of Natural Language Engineering on Statistical Learning of Natural Language Structured Input and Output