Information Retrieval

Terminology 🧸:

1. Document: A retrieval unit, over which a retreival system is built (ex. book chapter, a blog, a research paper)
2. Term: An indexed unit, usually words
3. Corpus/Collection: A group of documents over which retrieval is performed
4. Information need: A topic about which user desires to know more
5. Query: The one that user conveys to the computer in order to communicate the information need
6. Effectiveness: The quality of IR’s result(usually measured in precision and recall)

Information retreival is finding material(usually document) of an unstructured nature(usually text) that satifies an information neeed from within large collection. The example include web search, email search, grouping documents and many more.

Information could be categorized into three types. Structured(relations with well defiend attributes and values), Unstructured(free text), and Semi-structured(mixture of both).

Goal of Information retreival 🚀

Classic search model

The goal of the information retrieval is to retrieve documents with information that is relavant to the user’s information need and help the user complete a task. The quality of the retrieved document could be measure by precisions and recall which will be talked about in the later post.