Natural Language Processing - Unit I Notes (10 Marks Format)

Q1: Explain the study of language and its importance in NLP.

The study of language focuses on understanding how humans communicate through structured systems. It

lays the foundation for Natural Language Processing (NLP), which aims to replicate this understanding in

machines.

Key Points:

- Symbolic System: Language uses arbitrary symbols (words) to represent ideas.

- Rule-governed: Grammar dictates sentence structure and word order.

- Universality & Diversity: While all humans use language, each natural language has unique syntax and

semantics.

- Applications in NLP: A deep understanding helps in designing speech recognition, machine translation, and

sentiment analysis systems.

Importance in NLP:

- Helps in defining how machines interpret human language.

- Influences parsing, language modeling, and information extraction.

Conclusion: The study of language forms the theoretical base for building NLP models that can understand,

interpret, and generate human-like text.

Q2: Discuss the major applications of Natural Language Processing.

NLP bridges the gap between human communication and computer understanding. Its applications are

wide-ranging and deeply integrated into our digital lives.

Key Applications:

1. Machine Translation: Converts text from one language to another (e.g., Google Translate).

2. Speech Recognition: Converts spoken language into text (e.g., Alexa, Siri).

3. Text Summarization: Extracts essential information from large documents.

4. Sentiment Analysis: Determines the emotional tone in a text.

5. Information Retrieval: Improves search engine accuracy.

6. Question Answering & Chatbots: Interactive systems that understand and respond to user queries.
 Natural Language Processing - Unit I Notes (10 Marks Format)

Conclusion: NLP applications enhance user experience in communication, education, business, and

healthcare, making machines more human-aware.

Q3: How are Language Understanding Systems evaluated? Explain.

Evaluating NLP systems ensures their performance, accuracy, and efficiency in real-world applications.

Evaluation Methods:

- Precision: Correct outputs among all generated.

- Recall: Correct outputs among all relevant outputs.

- F1 Score: Harmonic mean of precision & recall.

- BLEU Score: Measures translation quality.

- Accuracy: Total correct predictions.

Intrinsic vs Extrinsic Evaluation:

- Intrinsic: Assesses isolated NLP components like POS tagging or parsing.

- Extrinsic: Evaluates overall system performance in real tasks.

Conclusion: Effective evaluation leads to robust NLP systems capable of performing reliably in various

applications.

Q4: Describe different levels of language analysis in NLP with examples.

Language understanding in NLP requires processing at multiple linguistic levels. Each level handles a

different aspect of human language.

Levels of Analysis:

- Phonology: Sound structure of words (e.g., 'Cat' vs 'Cut')

- Morphology: Word formation (e.g., 'Unbelievable' = Un + believe + able)

- Syntax: Grammatical structure of sentences (e.g., 'She eats an apple.')

- Semantics: Meaning of words and sentences (e.g., 'Bank' = river or finance)

- Pragmatics: Contextual use of language (e.g., 'Can you pass the salt?')
 Natural Language Processing - Unit I Notes (10 Marks Format)

- Discourse: Coherence across sentences (e.g., paragraph structure)

- World Knowledge: Understanding real-world facts (e.g., 'Ice is cold.')

Conclusion: Each level of language analysis contributes to deeper machine understanding, making NLP

systems more accurate and intelligent.

Q5: Explain the architecture of a Natural Language Understanding System.

Natural Language Understanding (NLU) systems analyze and interpret human language using structured

pipelines.

Components:

1. Tokenizer: Splits text into tokens (words, punctuation).

2. Part-of-Speech Tagger: Tags each word with its grammatical role.

3. Parser: Analyzes sentence structure (syntax trees).

4. Semantic Analyzer: Understands the meaning of words/sentences.

5. Discourse Analyzer: Connects meaning across sentences.

6. Inference Engine: Applies logic for understanding implications.

Diagram:

Input Text -> Tokenizer -> POS Tagger -> Parser -> Semantic Analysis -> Inference

Conclusion: NLU systems combine linguistic analysis and AI techniques to simulate human-like language

comprehension.

Q6: Write a short note on English syntax with an example.

Syntax governs the structure of sentences. English follows a specific syntactic order, and understanding it is

vital for parsing.

Key Concepts:

- Parts of Speech: Nouns, verbs, adjectives, etc.

- Phrase Structure Rules: Define how words combine:

S -> NP + VP

NP -> Det + N

VP -> V + NP

Parse Tree Example:

/ \

NP VP

/\ / \

Det N V NP

The dog ate food

Conclusion: English syntax provides the structure necessary for machines to understand and generate

grammatically correct sentences.