Information Retrieval Glossary

Information retrieval is vital to everyday life as technology takes on a larger role in society as a whole. Below is a glossary of terms that helps in understanding how information retrieval works.

• ASCII: Standard binary codes that represent occidental characters in a single byte.

• Ad hoc retrieval: A standard retrieval task where a user specifies an information need. This initiates a search by the system for any documents likely to be relevant to the query.

• Belief network: Model of data recovery that’s based on an interpretation of documents as well as nodes of a Bayesian network and index terms.

• Browsing: Interactive task where a user explores the web looking for documents that satisfy their information needs.

• CISI collection: Collection composed of 1460 documents that are selected from an ISI-assembled collection.

• Coding: Act of substituting text symbols with numeric codes that encrypt or compress text.

• Conversion: Changing of one format to a separate format.

• DLITE: Data recovery system that has two parts of functionality: control of a search process and a display of the search’s results.

• Data mining: Process of extracting new data or information from any other type of data.

• Data Recovery Service : Process of extracting data from damaged, failed, corrupted, or inaccessible storage media when it cannot be accessed normally.

• Digital library: Collection of digital objects. It refers to a system that preserves catalogue and archive information, allowing for browsing and retrieval by users.

• Directory: Hierarchical categorization of similar concepts in a certain domain of knowledge.

• Document: Specific unit of retrieval. It can be a Web page, article, book, section or chapter.

• Edit distance : Minimum number of replacements and insertions of characters required to make two strings equal to each other.

• Exact match: Mechanism where only objects matching specific criteria are returned as an answer.

• Extended Boolean model : Model of document recovery based on extensions of a classic Boolean model.

• Feature information: Information extracted from a specific object used during a query process.

• Filtering retrieval task: A task where a need for information by a user remains static as new documents enter a system.

• Fuzzy model: Theoretical model of document retrieval that’s based on a so-called “fuzzy-theory.”

• Global analysis: Reference technique of identifying term or document relationships through an analysis of all the other documents in the collection.

• Guided tour: Sequence of navigation choices that present the nodes in a specific order for some goal.

• Hard drive data recovery : Process of extracting data from damaged hard drive, failed hard drive, corrupted hard drive, or inaccessible hard drive storage media.

• Human-Computer Interfaces (HCI): Study of certain interfaces that assist a user with their information-seeking tasks like tracking of a retrieval disk or query formulation.

• Hypertext model: System of (info retrieval) based on the representation of certain document relationships as edges on a generic graph.

• Index term/keyword: Pre-selected term used to refer to specific content in a document. Used by common search engines.

• Inference network: Model of document retrieval that uses document interpretation, index terms, and user queries as nodes of a Bayesian network.

• Information retrieval: A part of computer science that studies methods of information retrieval from a collection of documents.

• Informative feedback: Information given to the user about any relationship between a query and the document retrieved.

• Inverted file index : Text index composed of a list of occurrence and a vocabulary.

• KWIC (KeyWords In Context): Technique that displays any occurrences of query terms within the context of documents being retrieved.

• Local analysis: A reference to techniques of identifying document and term relationships through the analysis of the documents retrieved by a given user query.

• Local context analysis: Method of query expansion that combines local and global analysis.

• MULTOS system: A document server with advanced multimedia retrieval capabilities.

• Metasearch: Technique used on the web where a point of entry is given to multiple heterogeneous back-end search engines.

• Multimedia data: Data that combines different media sources like text, images video or sound.

• Neural networks: Model of document retrieval that uses index terms, documents on a neutral network, and a representing query.

• OPAC (Online Public Access Catalogue): System of library management software that grants access to a library collection to interested users.

• Occurrence list: Data structure that assigns a list of positions to each text word within a document.

• Query: The expression of asking for the retrieval of desired information.

• Retrieval Task: Occurs when an information system executes a task in response to a specific user request.

• Raid Data Recovery : Process of extracting data from damaged raid, failed raid, corrupted raid, or inaccessible raid storage media when it cannot be accessed normally.

• SISD: Sequential computer architecture that consists of an instruction stream and data stream.

• Search history: Method of tracking a user’s session history

• Stemming : Technique that reduces words to their roots.

• Tag: String used to mark the end or beginning of elements in a text.

• User information need: Language declaration from users about their information needs.

• Vocabulary: Set of words in a text.

• XML: Subset of SGML which makes it easier to define markup languages.