Page 107 - Touhpad Ai

NLP in Internet Search Types
Natural Language Processing (NLP) plays a pivotal role in improving various types of internet searches by understanding
user intent, enhancing relevance, and enabling intelligent interaction with search systems.
Below are different types of Internet searches and how NLP enhances each:

Types of
Internet
Searches

Navigation Informational Transactional Investigative Voice
Searches Searches Searches Searches Searches

The descriptions of these types of Internet searches are as follows:
u Navigation searches: Navigation searches help users find a specific, known website or platform, often related to
popular brands or domains such as LinkedIn or YouTube. NLP identifies brand names, proper nouns, and phrases
indicating a user’s intent to navigate directly. Through entity recognition and query classification, the system accurately
directs users, prioritising official site links over generic content. For instance, searches like “YouTube login” or “LinkedIn
profile” will lead users straight to those platforms.
u Informational searches: Informational searches are made by users looking for knowledge or answers, without the
intent to make a transaction. NLP enhances these searches by applying semantic parsing, question-answering
models, and entity extraction to interpret the user's intent and deliver precise, contextually relevant information. For
example, queries like “What is artificial intelligence?” or “Symptoms of COVID” are understood as requests for factual
knowledge rather than commercial content.
u Transactional searches: Transactional searches indicate a user's intent to perform specific actions, such as
purchasing, downloading, or signing up for services. NLP detects action verbs like “buy,” “order,” or “download” in
conjunction with product or service names. This enables the search engine to provide relevant transactional results,
such as e-commerce pages or download links. Examples include searches like “Buy iPhone 15” or “Download Netflix
app,” which lead users directly to purchase or download options.
u Investigative searches: Investigative searches involve users comparing options, conducting detailed research, or
looking for reviews before making decisions. NLP identifies comparative terms such as “best,” “vs,” or “top-rated,”
and extracts sentiments and feature summaries from reviews or articles. Semantic search techniques help organise
content for easier comparison, supporting queries like “Best web series 2025” or “MBA vs M.Tech career scope,”
yielding rankings, expert opinions, and detailed reviews.
u Voice searches: Voice searches are made using spoken language, typically with devices like Siri, Alexa, or
smartphone assistants. These queries are often more conversational and natural. NLP first converts speech to text
through automatic speech recognition (ASR), then applies natural language understanding (NLU) to decode user
intent, context, and entities. Voice searches are capable of handling less structured, sometimes ambiguous language,
enabling fast, contextual responses to commands like “Is it going to rain today?” or “Play relaxing music.”

Potential Use of AI
Artificial Intelligence (AI) has revolutionised multiple sectors, automating processes, improving efficiency, and
enhancing user experiences.

Introduction and State of Art of AI, Natural Language Processing (NLP), and Potential use of AI 105