• Ryan McDonald and Joakim Nivre‘s course on dependency parsing.
• 2006 CoNLL-X Shared Task on Dependency Parsing
• 2007 CoNLL Shared Task on Dependency Parsing

• MSTParser
• MaltParser

Resources:

• PropBank, Palmer et al, 2005: The proposition bank: An annotated corpus of semantic roles.
• FrameNet, Baker, Fillmore and Lowe, 1998: The Berkeley FrameNet project
• NomBank
• VerbNet
• Salsa
• Ellsworth et al, 2004: PropBank, SALSA, and FrameNet: How design determines product. A paper comparing PropBank, FrameNet and Salsa

Papers on semantic role labeling:

• Gildea and Jurafsky 2002: Automatic Labeling Of Semantic Roles The first paper describing SRL, basic feature set, relatively complex learning paradigm
• Xue and Palmer, 2004: Calibrating Features for Semantic Role Labeling This paper describes a pruning scheme that is widely used today.
• Surdeanu et al 2003: Using Predicate-Argument Structures for Information Extraction, Pradhan et al 2004: Shallow Semantic Parsing using Support Vector Machines The feature sets used in these two papers are becoming a quasi-standard.
• Pradhan, Ward and Martin 2007: Towards Robust Semantic Role Labeling A paper on the problem of domain transfer in SRL.

• ASSERT
• SwiRL
• Shalmaneser

• Shared Task Website

• Chris Manning’s lecture on unsupervised parsing and semantic role labeling.

• WordNet-based methods: Resnik, Cognition 96, Li and Abe, CL 98, Clark and Weir, CL 02, Ciaramita and Johnson, COLING 00, and an overview paper:Brockmann and Lapata, EACL 03
• Clustering methods: Rooth et al, ACL 99

• Look at last year's L&C course to see what was covered then and grab some of the resources.
• Search for HAL 9000 on YouTube to look at some clips from 2001: A Space Odyssey.
• Check out the webpage for the course on lego robots at Saarland University. See videos of the lego robot bartender we saw in class.
• Check out the CoSy webpage, which is the project that works on the mobile robot we saw a couple of videos of in class.
• See the How to say NO to a robot video.
• Watch the humanoid robocup final that we saw the start of in class.
• Watch the first soccer match between man and machine.
• Check out the Geekipedia by WIRED magazine.
• Learn everything you ever wanted to know about leet and pwn.
• Learn about Lolcats. You can even read the lolcat Bible.

• Browse the world of writing systems at Omniglot.
• Get to know the International Phonetic Alphabet.
• Language Log explains what is great about Hangul.
• Read about the Tengwar writing system Tolkein devised for languages of Middle Earth.

• Text-to-speech systems: IBM and AT&T
• Hear the Radiolab episode about music and language. (We listened to the segment on Chinese tones.)
• Check out a tutorial on speech.
• See a page containing 2D and waterfall spectrograms for the syllable 'ka'.

• The UT Austin library advanced search page
• Advanced search: here’s the Google help cheat sheet
• Search the British National Corpus
• Search the European Parliament Corpus
• Try out the regular expression exercises from Lexmasterclass.
• The Wikipedia regular expressions page: good for looking up language constructs
• Information about regular expressions

• Spam and anti-spam: What Nigeria spam is and how people have tricked the tricksters
• The infinite monkey theorem and monkeys typing made concrete.
• Scientific American: Stopping Spam, by Joshua Goodman, David Heckerman and Robert Rounthwaite.

• The Wikipedia page on question answering has a nice overview.
• Edinburgh’s QuALiM question answering system (uses Wikipedia).
• ISI’s Textmap System. In particular, check out the question representation after you ask a question.
• MIT CSAIL’s START QA system.
• Wikipedia: presuppositions.

• Talk to ELIZA.
• Talk to A.L.I.C.E.
• Talk to Jabberwacky chatbots.
• Jabberwacky chats with A.L.I.C.E – quite amusing!
• Terry Winograd’s page about the SHRDLU system.
• Watch the Saturday Night Live skit about a blind date with a dialog system.
• A big list of links related to dialog systems.
• The COMIC project: Conversational Multimodal Interaction with Computers.
• The CoSy project: dialog systems and cognitive architectures for robots.
• Check out the webpage for the Cognitive Machines group at the MIT media lab.
• The WITAS project: multi-modal conversational interfaces (including autonomous (simulated) helicopter).
• The NICE project: dialog system for edutainment.
• Wikipedia: the Chinese room.
• Terry Brisson’s meaty reply to the Chinese room.
• Radiolab episode on emergence.
• Wikipedia: emergence.
• Wikipedia: homunculus.
• Tux Droid: talking penguin.

• Peter Norvig has written a nice article about spelling correction, using Python to make an implementation very quickly.
• If you need some help with data structures, look at How to Think Like a Computer Scientist: Learning with Python.

• Wikipedia: Cryptography Portal.
• Wikipedia: the Vigenere cipher.
• The Code Book website.
• The cipher of Mary Queen of Scots.
• Wikipedia: the Enigma machine used by the Germans in WWII.
• Wikipedia: The Navajo code talkers who provided secure communication for the USA during WWII.
• Wikipedia: The Rosetta Stone, which provided the key to deciphering Egyptian hieroglyphs.
• Wikipedia: the Voynich manuscript.
• Wikipedia: Zipf's law.

• Try out Google's online machine translation system.

• Have a look at touch screens for manipulating images and data structures. Imagine search results looking like that! [Thanks to Edmund Lee for the link.]
• The Urban Dictionary has a lot of great new words, or new meanings for old words. Great for multislacking!

• Try using Google Sets to expand on a set of related words.
• Computers are being used to play music.

• It can be difficult to say no to a robot... (warning: explicit language) Or, to put it differently: Today’s speech recognition systems get in trouble when the utterance is short.

• Learn more about how computers are being used to read lips for a variety of applications.
• A cool video about how the Machine is us/ing us.

• Try the word association game or JeuxDeMots.
• Have ESP?

How about some recreational corpus lingistics? Language Log has some fun studies, particularly in Googlometry:

• Nearly no? Almost no? with followups here, here with an introspective analysis, and here
• "Boxes sitting around still unpacked": anything weird about that phrase? with followups here, here, here, and here, with musings on prescriptivism and descriptivism
• Is "far from" becoming an adverb?, with a great discussion of the reliability of Google counts
• Half full or half empty, and more on Google counts
• And apropos of the reliability of Google counts...
• Do women really produce more words than men? Let’s ask a corpus.
• The shape of a spoken phrase
• Google counts and French politics, and other corpus studies by Jean Veronis
• In the circumstances? under? within??
• Variations over a phrase: I am X, hear me Y
• As he is known as?
• Would you like fewer politics, or rather fewer physics?, with followup
• "Why are negations so easy to fail to miss?"
• Gone fishing? Gone it alone?

Search tools:

• The Linguist's search engine: Search syntactic analyses. You can prune a syntactic tree graphically to construct your query – very convenient!
• tregex: a powerful tool for searching syntactically analyzed corpora (though the query language is quite complex)
• TIGERSearch: another powerful tool for searching syntactically analyzed corpora, with graphical display of search results
• The IMS corpus workbench

Online corpora with search facility:

• OPUS: a collection of parallel corpora, searchable with cqp queries

Finite state automata:

• Get the Visual Automata Simulator

Textual Entailment:

• The RTE website has info on all three challenges, as well as the data for download.
• The data from RTE-1 and RTE-2 is also available at /groups/corpora/RTE in the computational linguistics lab.

NLP in the news:

• Exploiting voice-recognition to do bad things in Windows Vista.

NLP links:

• The Natural Language Toolkit.
• The NLTK Tutorials.
• The Finite State Morphology homepage for the Beesley and Karttunen book and the Xerox Finite State Toolkit (XFST).
• More semantics: view the lecture notes of the Semantic Theory course at the Saarbruecken Computational Linguistics Institute
• The Association for Computational Linguistics.

Python Readings:

• The NLTK Python tutorial chapter.
• The Python Tutorial.
• Python for Linguists.

Keeping a lab book for chronicling experiments:

• Journler looks like an interesting tool for this.

• Lexicon resources:
  • WordNet
  • FrameNet
  • The Levin class index
  • The Unified Verb Index combines FrameNet, PropBank, and VerbNet
  • Lin's thesaurus was constructed automatically using a mutual information-based semantic similarity metric
• The Word sketch engine shows typical contexts of a word, to aid lexicographers
• SIGLEX, the ACL special interest group on the lexicon, has lists of online resources

• The Noun Associations for German searchable database of free associations for concrete German nouns

• Here’s a great resource containing data sets from all sorts of psychological and psycholinguistic studies, including feature norms and such things: The Psychonomic Society Archive of Norms, Stimuli, and Data

• the CCG Site has papers and links related to Combinatory Categorial Grammar
• OpenCCG is a non-probabalistic CCG parsing system that can do both parsing and realization
• the OpenCCG web frontend developed right here at UT Austin
• There is also a tutorial for the OpenCCG GUI and a guide to grammar writing with OpenCCG.
• the C&C CCG parser and toolkit is a fast statistical CCG parser with lots of other goodies (eg, POS tagger and chunker)
• the old CG homepage, though not sure what it is doing in China!