Information Retrieval: Algorithms and Heuristics by David A. Grossman

By David A. Grossman

Information Retrieval: Algorithms and Heuristics is a finished creation to the examine of data retrieval masking either effectiveness and run-time functionality. the point of interest of the presentation is on algorithms and heuristics used to discover files appropriate to the person request and to discover them quick. via a number of examples, the main established algorithms and heuristics wanted are tackled. To facilitate figuring out and purposes, introductions to and discussions of computational linguistics, average language processing, likelihood concept and library and desktop technology are supplied. whereas this article specializes in algorithms and never on advertisement product according to se, the fundamental suggestions utilized by many advertisement items are defined. strategies that may be used to discover details on the net, in addition to in different huge info collections, are integrated.
This quantity is a useful source for researchers, practitioners, and scholars operating in info retrieval and databases. For teachers, a collection of Powerpoint slides, together with speaker notes, can be found on-line from the authors.

Show description

Read or Download Information Retrieval: Algorithms and Heuristics PDF

Similar desktop publishing books

PostScript Language Tutorial and Cookbook

PostScript, a innovative, device-independent web page description language, is instantly changing into the average for printing high quality built-in textual content and portraits. it's a robust, versatile language that has the power to explain successfully the looks of textual content, photos, and picture fabric at the revealed web page.

Italian Music Incunabula: Printers and Type

Musical notation offered strange demanding situations to the hot craft of printing within the 15th century. Its calls for have been so tricky that the 1st impact of track from steel style used to be no longer made till a whole two decades after the 1st published alphabetic texts. by way of the top of the century dozens of such fonts had seemed all through Europe.

Manly Crafts

Manly Crafts offers complete step by step directions for 35 artful initiatives with not anything frilly, lacy, or pastel in sight. the right way to make a 5-cent marriage ceremony band, leather-based iPod case, and paracord bullwhip!

Additional resources for Information Retrieval: Algorithms and Heuristics

Example text

If the whole document is used as a component, then we are back at traditional probabilistic information retrieval. The basic weight for a given component is defined as the ratio of the probability of the component being relevant to the probability that it is not relevant. This is- RETRIEVAL STRATEGIES Wale role and Sale = In ( role ) role) (1- 41 + In (1- Sale)) Sale are weights that can be estimated in one of three different ways- • Initial estimate using self-relevance. A component which is relevant to itself, results in- where La is the number of components in the document, and Fie is the number of occurrences of term k in the collection.

Term frequency was not used in the original probabilistic model. Croft and Harper incorporate term frequency weights in [Croft and Harper, 1979]. Relevance is estimated by including the probability that a term will appear in a given document, rather than the simple presence or absence of a term in a document. The term frequency is used to derive an estimate of how likely it is for the term to appear in a document. This new coefficient is given below. The P(d;j) indicates the probability that term i appears in document j, and can be estimated simply as the term frequency of term i in document j.

For any given game, there is a seventy five percent chance that the team will win if the weather is sunny and a sixty percent chance that the team will win if the shortstop plays. 6 The conditional probability that the team will win given both situations is written as p(win I sunny, good-shortstop). " We have two pieces of evidence indicating that the Salamanders will win. Intuition says that together the two pieces should be stronger than either alone. " A seventy-five percent chance of winning is a twenty-five percent chance of losing, and a sixty percent chance of winning is a forty percent chance of losing.

Download PDF sample

Rated 4.45 of 5 – based on 42 votes