AI & Chatbot Glossary

RAG is an AI architecture that retrieves relevant documents from a knowledge base before generating a response, grounding LLM output in verified facts.

NLP is the branch of artificial intelligence that gives machines the ability to read, understand, and generate human language.

Conversational AI is the set of technologies that enable machines to engage in natural, human-like dialogue through text or voice.

A chatbot widget is an embeddable user interface component that adds a conversational AI chat window to any website or application.

AI hallucination occurs when a language model generates plausible-sounding but factually incorrect or fabricated information with apparent confidence.

An AI knowledge base is a structured repository of content that an AI chatbot searches through to provide accurate, grounded answers to user questions.

Chatbot training is the process of providing a chatbot with the information and instructions it needs to accurately answer user questions and carry out tasks.

Intent detection is the NLP task of classifying a user message into a predefined category that represents what the user is trying to accomplish.

Sentiment analysis is the NLP technique that identifies the emotional tone of text, classifying it as positive, negative, neutral, or more granular emotional states.

Live chat handoff is the process of seamlessly transferring a chatbot conversation to a human support agent, preserving the full conversation context.

Omnichannel messaging is a unified communication strategy that connects all customer messaging channels into a single inbox with shared context and conversation history.

A webhook is an HTTP callback that automatically sends data from one system to another in real time when a specific event occurs.

An AI embedding is a dense numerical vector that represents the semantic meaning of a piece of text, enabling mathematical comparison of concepts.

A vector database is a specialized data store optimized for indexing, storing, and querying high-dimensional embedding vectors using similarity search algorithms.

Prompt engineering is the practice of designing and optimizing the text instructions given to an AI model to achieve desired behavior, tone, and output quality.

Chatbot analytics is the measurement and analysis of chatbot performance metrics including resolution rates, confidence scores, satisfaction, and conversation patterns.

A no-code chatbot builder is a platform that lets you create, train, and deploy AI chatbots through visual interfaces without writing any programming code.

Confidence scoring is the process of evaluating how certain an AI system is about its response, typically based on the quality and relevance of retrieved information.

Hybrid search combines dense vector search (semantic similarity) with sparse keyword search (exact matching) to deliver more accurate and comprehensive retrieval results.

Chunking is the process of splitting documents into smaller, semantically coherent segments that can be independently embedded and retrieved in a RAG pipeline.

Reranking is a second-pass scoring step that uses a more powerful model to re-evaluate and reorder initial search results for greater precision.

A system prompt is a set of hidden instructions provided to an AI model that defines its role, personality, constraints, and behavior throughout a conversation.

A multilingual chatbot automatically detects the language of incoming messages and responds naturally in that language, serving global audiences without manual translation.

A chatbot embed code is a small JavaScript snippet that website owners paste into their HTML to add an AI chatbot widget to their site.

Conversational commerce is the practice of using messaging channels and AI chatbots to guide customers through the buying journey, from product discovery to purchase.