Natural Language Processing

Natural Language Processing, or NLP, is the area of artificial intelligence that enables computers to work with human language in written or spoken form. It includes tasks such as text classification, translation, speech recognition, question answering, and conversational interaction. Manufacturing hardware, studying plants, and printing images are unrelated activities, so they do not describe the central purpose of NLP.

Sentiment analysis is an NLP task that examines words, phrases, or complete texts to infer the writer’s opinion, attitude, or emotional tone. A basic system may classify a review as positive, negative, or neutral, while advanced systems can identify particular emotions or aspects. Grammar correction, machine translation, and part-of-speech tagging are useful NLP tasks, but they answer different questions.

Stop words are very common function words, such as “the,” “and,” or “is,” that may contribute limited information for particular NLP tasks. Removing them can reduce vocabulary size, lower processing effort, and help a model emphasize more informative terms. Option A is correct, but removal is task-dependent because some stop words can matter for meaning, sentiment, or grammar. It does not remove every important word.

Text classification assigns written content to predefined labels, often by learning patterns in words or other language features. An email-spam filter examines the message and assigns it to categories such as spam or legitimate, so option A is correct. Resizing an image is an image-processing operation, changing audio volume is sound processing, and making a chair is a physical manufacturing activity rather than a language task.

Named Entity Recognition is an NLP information-extraction technique that locates and classifies specific entities in text, such as people, places, organizations, dates, or amounts. For example, it can label “Delhi” as a location and “Ravi” as a person. Sentiment analysis studies emotions, translation changes languages, and summarization shortens content, so option A is correct.

Context helps an NLP system determine the intended meaning of words and sentences. For example, “bank” may mean a financial institution or the side of a river, depending on nearby words and the situation. Context supports more than spelling correction, but it does not completely replace grammatical rules.

Machine translation is an NLP application that automatically converts text or speech from one human language into another, such as translating a Hindi sentence into English. It may use rules, statistical methods, or neural models, and the quality can depend on context and language pair. Image resizing, shutting down a computer, and lowering sound volume are unrelated operations, so option A is correct.

Speech-to-text, also called automatic speech recognition, processes an audio signal, detects spoken words, and produces written or digital text. Therefore, option A is correct because it describes conversion from speech to text. Option B is the reverse process, text-to-speech. Option C belongs to computer vision, while option D is a text-editing or language-processing task.

Sentiment analysis is an NLP technique that examines the opinion or emotional tone expressed in text and classifies a review as positive, negative, or neutral. Therefore, option A is correct. Converting text into an image, deleting a customer’s address, and reversing word order are unrelated operations. Exam tip: Choose sentiment analysis when the task is to classify the opinion or emotional tone of a review.

Many words have more than one meaning. For example, “bank” may refer to a riverbank or a financial institution. The sentence and situation provide the context needed to identify the intended meaning, making machine translation more accurate and natural. Therefore, option A is correct; option B is incorrect because many words are ambiguous. Exam tip: Use the surrounding context to determine the meaning of a word during translation.

Speech recognition processes an audio signal containing human speech. A microphone or another audio sensor captures the sound, and the system analyses speech patterns to convert them into words or text; therefore option A is correct. Numerical tables are structured data, printed documents require document or optical-character recognition, and still images are handled by computer-vision systems.