IRS_in的troduction.pptVIP

BACK to IRS Syllabus Lecture One, Introduction What is Information Retrieval (IR) The Components of IRS Glossary for IR What’s IR Finding relevant information in large collections of data In such a collection you may want to find: “Give me information on the history of the DL” An article about the Digital Library (text retrieval) “What does a brain tumor look like on a CT-scan” A picture of a brain tumor (image retrieval) “It goes like this: hmm hmm hahmmm…” A certain song (music retrieval) What’s IR (Cont.) Definition A research with a goal of exploration of information storage, classification, extraction, indexing and browsing techniques for the retrieval of non-structural databases such as textual documents What’s IR (Cont.) IR is the study of unstructured data Usually text, but also audio, images, protein sequences, … A document is a container for data with unknown structure Document format usually is known Example: HTML, XML, Word, … Document structure may or may not be known Example: Where titles begin/end, where section begin/end What’s IR (Cont.) Structure of underlying data is not known Example: Syntax and semantics of text in each section Types of Documents Text (default), hypertext, image, video, multimedia, … IR systems exploit statistical regularities in the data Without trying to “understand” what they mean Why is IR important Most communication between humans is unstructured information Text, images, speech, audio It is common to store information in electronic form Word processing systems have been common Storage devices have become very inexpensive The rate of information production is growing rapidly, but … People have less time to organize it for easy access People have less time to spend looking for what they need Why is IR important (Cont.) The Internet provides easy access to information worldwide, but … The information may be disorganized, or organized for other uses The information you need may be in language you don’t speak IR vs RDBMS/NL