Lecture 1 Flashcards
Introduction to NLP & Levels of Language
NLP - Linguistics
Formal, structural models of language
NLP - Computer Science
Internal representations of data and algorithms for efficient processing
NLP - Artificial Intelligence
Computational theory of human language processing
NLP - Cognitive Psychology
Human cognition in language
NLP - Statistics
Frequencies and probabilities of linguistic patterns
Computational Linguistics
doing linguistics on computers
Natural Language Processing (NLP)
A range of computational
techniques:
* for analyzing and representing
naturally occurring texts at various
levels of linguistic analysis
* for the purpose of achieving
human like language processing
* for a range of particular tasks or
applications
NLP Application Areas:
Machine Translation
conversion of text from one language to another
NLP Application Areas:
Information Extraction
populating a database with specific data found in text
NLP Application Areas:
Human computer Interfaces
NLP assistants, chatbots , interactive querying of databases
NLP Application Areas:
Summarization
abstraction and condensation of major points
NLP Application Areas:
Question Answering Systems
provision of best answer to the given question
NLP Application Areas:
Information Retrieval / Search Engines
provision of documents
NLP Application Areas:
Image to Text, Speech to Text and
Text to speech solutions
Support
Data:
Unstructured
Data that has no inherent structure and is usually stored as different types of files
eg txt, pdf, images, videos
Data:
Quasi-Structured
Textual Data with erratic formats that can be formatted with effort and software tools
eg JSON (JavaScript Object Notation)
Data:
Semi-Structured
Textual data files with an apparent pattern, enabling analysis
eg spreadsheets and XML files
Data:
Structured
Data having a defined data model, format, structure
eg database
Why is NLP difficult?
Usually people are largely
unaware of the complexity of the
language tasks they perform so
effortlessly (exception: adult
Why is NLP difficult?
- Subtleties of meaning
- Irony, sarcasm, humor, metaphor
- Ambiguity
- Ambiguity is a fundamental
problem of computational
linguistics - Resolving ambiguity is a crucial goal
Levels of Language Analysis (1)
Phonetic
Levels of Language Analysis (2)
Morphological
Levels of Language Analysis (3)
Lexical
Levels of Language Analysis (4)
Syntactic
Levels of Language Analysis (5)
Symantec
Levels of Language Analysis (6)
Discourse
Levels of Language Analysis (7)
Pragmatic
Transitive:
eg Duck
(verb has a noun direct object)
eg I cooked [waterfowl belonging to her]
Ditransitive:
eg Duck
(verb has 2 noun objects)
eg I made [her] (into) [undifferentiated waterfowl]
Action-transitive
eg Duck
(verb has a direct object and another verb)
eg I caused [her] [to move her body]
Ambiguity at Pragmatic Level
eg ‘At what time did you call me?’
Literal Meaning: What time was it?
Literal Response: A time (e.g. ’11 am.’)
Ambiguity at Pragmatic Level
eg ‘At what time did you call me?’
(Pragmatic Meaning: a different question entirely,
e.g. Why were you so late?
Pragmatic Response: Explain the reason for being so late.)
Natural Language Toolkit (NLTK)
A suite of Python libraries for symbolic and statistical
natural language programming
* Developed at the University of Pennsylvania
