(SEM VIII) THEORY EXAMINATION 2022-23 NATURAL LANGUAGE PROCESSING

NATURAL LANGUAGE PROCESSING (KOE-088)

B.Tech Semester VIII – Theory Answers

SECTION A

(a) Pragmatic analysis

Pragmatic analysis deals with understanding the intended meaning of a sentence based on context rather than just its literal meaning. It considers speaker intention, real-world knowledge, and situational context to interpret language correctly. Pragmatics helps resolve ambiguities and understand implied meanings, sarcasm, and indirect requests in natural language communication.

(b) Discourse integration

Discourse integration refers to the process of connecting individual sentences to form a coherent and meaningful whole. It involves maintaining context, resolving references such as pronouns, and understanding relationships between sentences. This process ensures continuity and consistency in conversations or written text.

(c) Basic moods of sentences

The basic moods of sentences express the speaker’s intention. Declarative sentences state facts, interrogative sentences ask questions, imperative sentences give commands or requests, and exclamatory sentences express strong emotions. Identifying sentence mood helps in correct interpretation and processing.

(d) Meta-knowledge related to AI systems

Meta-knowledge is knowledge about knowledge itself. In AI systems, it includes information about how knowledge is structured, accessed, used, and modified. Meta-knowledge helps systems reason efficiently, choose appropriate problem-solving strategies, and manage uncertainty.

(e) Top-down parser

A top-down parser starts parsing from the start symbol of the grammar and attempts to generate the input sentence by expanding grammar rules. It predicts sentence structure before reading all input symbols. While efficient for simple grammars, it may suffer from backtracking issues.

(f) Morphological analysis

Morphological analysis studies the internal structure of words. It identifies roots, prefixes, suffixes, and inflectional forms. This analysis helps determine word meaning, grammatical role, and relationship with other words, which is essential for language understanding.

(g) Auxiliary verb with example

Auxiliary verbs assist main verbs in forming tense, voice, or mood. Examples include is, are, have, has, will, and can. For example, in the sentence “She is reading,” the verb is acts as an auxiliary.

(h) Deterministic parser

A deterministic parser follows a fixed set of rules to parse sentences without backtracking. It makes parsing decisions at each step based on current input. Such parsers are fast but may fail when faced with ambiguous grammar.

(i) Parts of speech tagging

Parts of speech tagging is the process of assigning grammatical categories such as noun, verb, adjective, or adverb to each word in a sentence. It helps in syntactic analysis and improves the accuracy of language understanding tasks.

(j) Encoding ambiguity in logical form

Encoding ambiguity in logical form occurs when a sentence has multiple interpretations due to structural or semantic ambiguity. Logical representations must capture these alternative meanings so that the system can choose the correct interpretation using context.

SECTION B

2(a) Applications of Natural Language Processing

Natural Language Processing is widely used in applications such as machine translation, speech recognition, chatbots, sentiment analysis, text summarization, and information retrieval. NLP enables machines to interact with humans in natural language, making systems more intelligent and user-friendly. These applications are crucial in areas like healthcare, education, customer support, and data analytics.

2(b) Techniques of knowledge representation

Knowledge representation techniques store and organize information in a form that machines can process. Common techniques include semantic networks, frames, logic-based representation, production rules, and scripts. These methods help AI systems reason, infer new facts, and solve problems efficiently.

2(c) Bottom-up chart parsing algorithm

Bottom-up chart parsing starts from the input symbols and builds parse trees by combining smaller constituents into larger ones. The chart stores intermediate parsing results to avoid redundant computation. This method efficiently handles ambiguity and improves parsing performance.

2(d) Handling questions in context-free grammar

Handling questions in context-free grammar involves transforming interrogative sentences into structured representations. Rules are defined to manage question words, word order changes, and auxiliary verbs. This allows NLP systems to understand and respond accurately to user queries.

2(e) Best-first parsing

Best-first parsing selects the most promising parse at each step based on heuristic evaluation. It prioritizes likely parses and explores alternatives only when necessary. This approach balances efficiency and accuracy in parsing complex sentences.

SECTION C

3(a) Steps in natural language understanding

Natural language understanding involves several steps starting with lexical analysis, where words are identified and categorized. This is followed by syntactic analysis to determine sentence structure. Semantic analysis assigns meaning, while discourse and pragmatic analysis interpret context and intention. Together, these steps enable machines to understand human language accurately.

3(b) Natural Language Processing and its components

Natural Language Processing is a field of AI that enables machines to process human language. Its main components include lexical processing, syntactic parsing, semantic interpretation, discourse analysis, and pragmatic understanding. These components work together to convert raw text into meaningful representations for intelligent applications.