21CSE552T

Computational Linguistics

Professional Elective (E)3 L0 T0 P3 CSchool of ComputingPrerequisites: Nil

An introduction to computational linguistics: the science behind natural language processing. The course builds from the role of NLP and the structure of linguistics, through grammar formalisms after Chomsky, the practical products of the field, the meaning to text view of language, and the linguistic models that underpin modern language technology.

Primary text: Igor A. Bolshakov and Alexander Gelbukh, Computational Linguistics: Models, Resources, Applications (IPN, 2004).

Course outcomes

By the end of the course, students will be able to:

Summarise the core concepts of computational linguistics

Construct and explain the various applications of computers in linguistics and language studies

Design and reason about tools for linguistic analysis

Apply the text transformation view of language to strengthen NLP systems

Apply the different modelling techniques based on linguistics

Units

The full plan for the course. Units with lectures published are linked.

Unit 1

Introduction

9 hours

The role of natural language processing. Linguistics and its structure. What we mean by computational linguistics. The important role of the fundamental science. Current state of applied research on Spanish.

View 6 lectures

Unit 2

Overview of Grammar

9 hours

A historical outline. The structuralist approach. Initial contribution of Chomsky. A simple context-free grammar. Transformational grammars. Linguistic research after Chomsky: valencies and interpretation, and the freedom of grammar to allow ungrammatical sentences. Linguistic research after Chomsky: constraints. Head-Driven Phrase Structure Grammar. The idea of unification. The Meaning-Text Theory: multistage transformer and government patterns. Dependency trees. Semantic links.

View 8 lectures

Unit 3

Products of Computational Linguistics

9 hours

Present and prospective products. Classification of applied linguistic systems. Automatic hyphenation. Spell checking. Grammar checking. Style checking. References to words and word combinations. Information retrieval. Topical summarization. Automatic translation. Natural language interface. Extraction of factual data from texts. Text generation. Systems of language understanding. Related systems.

View 8 lectures

Unit 4

Language as a Meaning to Text Transformer

9 hours

Text transformer. Possible points of view on natural language. Language as a bi-directional transformer. Two ways to represent meaning. Decomposition and atomization of meaning. Not-uniqueness of meaning to text mapping: synonymy. Not-uniqueness of text to meaning mapping: homonymy. More on homonymy. Multistage character of the meaning to text transformer. Translation as a multistage transformation. Two sides of a sign. The linguistic sign in the MMT and in HPSG. Generative, MTT, and constraint ideas in comparison. Case study: writing simple parsers in groups for regional languages.

Coming soon

Unit 5

Linguistic Models

9 hours

What is modeling in general. Neurolinguistic models. Psycholinguistic models. Functional models of language. Research linguistic models. Common features of modern models of language. Specific features of the meaning. Text model. Reduced models. Analogy in natural languages. Empirical versus rationalist approaches. Limited scope of the modern linguistic theories. Case study: applications involving language models and demonstration of simple application specific modules using tools.

Coming soon