In the context of information retrieval (IR), a document refers to any unit of information that is stored in a collection or database, such as a web page, an academic paper, an image, or a video. A document is typically the entity that the IR system searches through in response to a user query.
Documents can vary in structure and content; for example, they can be text-based (like articles or blog posts) or multimedia (like images or videos).
The goal of an IR system is to retrieve documents that match the user's query based on their content and relevance to the query.