Académique Documents
Professionnel Documents
Culture Documents
Pierre Nugues
Lund University Pierre.Nugues@cs.lth.se http://www.cs.lth.se/home/Pierre_Nugues/
Pierre Nugues
1 / 19
Spelling and grammatical checkers: MS Word Text indexing and information retrieval on the Internet: Google, Microsoft Bing, Yahoo Telephone information that understands some spoken questions: SJ (trains in Sweden) or Tellme.com in the United States Speech dictation of letters or reports: IBM ViaVoice, Windows Vista Translation: Google Translate, SYSTRAN
Pierre Nugues
2 / 19
Direct translation from spoken English to spoken Swedish in a restricted domain: SRI and SICS Voice control of domestic devices such as tape recorders: Philips or disc changers: MS Persona Conversational agents able to dialogue and to plan: TRAINS Spoken navigation in virtual worlds: Ulysse, Higgins Generation of 3D scenes from text: Carsim
Pierre Nugues
3 / 19
Linguistics Layers
Sounds Phonemes Words and morphology Syntax and functions Semantics Dialogue
Pierre Nugues
4 / 19
Serious
Pierre Nugues
5 / 19
The big cat ate the gray mouse The /article big /adjective cat /noun ate /verb the /article gray /adjective mouse /noun Le /article gros /adjectif chat /nom mange /verbe la/article souris /nom grise /adjectif Die /Artikel groe /Adjektiv Katze /Substantiv it /Verb die /Artikel graue /Adjektiv Maus /Substantiv
Pierre Nugues
6 / 19
Morphology
Root form to work + verb + preterit travailler + verb + past participle arbeiten + verb + past participle
Pierre Nugues
7 / 19
Syntactic Tree
sentence
noun phrase
verb phrase
article
noun
verb
noun phrase
article
noun
The
boy
hit
the
ball
Pierre Nugues
8 / 19
Pierre Nugues
9 / 19
Semantics
As opposed to syntax:
1 2
Colorless green ideas sleep furiously. *Furiously sleep ideas green colorless.
Determining the logical form: Sentence Frank is writing notes Fran cois ecrit des notes Franz schreibt Notizen Logical representation writing(Frank, notes). ecrit(Fran cois, notes). schreibt(Franz, Notizen).
Pierre Nugues
10 / 19
Lexical Semantics
Word senses:
1 2 3 4 5
note (noun) short piece of writing; note (noun) a single sound at a particular level; note (noun) a piece of paper money; note (verb ) to take notice of; note (noun) of note: of importance.
Pierre Nugues
11 / 19
Reference
3. real world
referencing
referencing
Pierre Nugues
12 / 19
Ambiguity
Many analyses are ambiguous. It makes language processing dicult. Ambiguity occurs in any layer: speech recognition, part-of-speech tagging, parsing, etc. Example of an ambiguous phonetic transcription: The boys eat the sandwiches That may correspond to: The boy seat the sandwiches ; the boy seat this and which is ; the buoys eat the sand which is
Pierre Nugues
13 / 19
Linguistics has produced an impressive set of theories and models Language processing requires signicant resources Models and tools have matured. Resources are available. Tools involve notably nite-state automata, regular expressions, rewriting rules, logic, statistics and machine learning.
Pierre Nugues
14 / 19
= NLP engine
Pierre Nugues
15 / 19
Pierre Nugues
16 / 19
Pierre Nugues
17 / 19
Pierre Nugues
18 / 19
Research Relevance
Large companies like Microsoft, Google, Yahoo, IBM, or Xerox have a research activity in natural language processing. The 7th European framework program (2007-2013) names six technology pillars in information technologies. Two of them are related to language processing: Knowledge, cognitive and learning systems: semantic systems; capturing and exploiting knowledge embedded in web and multimedia content; bio-inspired articial systems that perceive, understand, learn and evolve, and act autonomously; learning by convivial machines and humans based on a better understanding of human cognition. Simulation, visualization, interaction and mixed realities: tools for innovative design and creativity in products, services and digital media, and for natural, language-enabled and context-rich interaction and communication.
Pierre Nugues
19 / 19