Six search methods to retrieve documents from knowledge bases

A recent article describes six “search technologies,” each of which can help a law department’s members find useful material stored on drives, intranets, or databases. The ABA J., Vol. 95, April 2009 at 43, focuses on litigation support rather than knowledge management, but it discusses these ways to search through text and summarizes them in a sidebar (See my post of Feb. 23, 2006: four patents for search methodologies; Dec. 10, 2005: implicit search software; June 30, 2006: new search tools; and July 21, 2005: software that helps collect knowledge.).

Boolean search, using terms along with connecting words like “and” and “or” (See my post of March 5, 2005: Google Desktop Search.).

Fuzzy search, that extends the Boolean search beyond the exact term entered, such as “general counsels” if someone enters “general counsel” (See my post of March 23, 2006: knowledge management software and fuzzy logic.).

Algebraic search, which is “based on the premise that mathematical models can figure out the meaning of a document and then retrieve relevant documents by looking at the proximity of related words” (See my post of Dec. 10, 2005: implicit search software.).

Probabilistic search “uses language models, including Bayesian belief networks” (See my post of May 24, 2005: analysis searching software.).

Concept search, which often involve complex mathematical and linguistic models, “but usually take human training to ‘teach’ the computers to recognize terms and concepts. Sometimes this involves development of a thesaurus of search terms and variations (See my post of Oct. 31, 2005: thesaurus searching; May 24, 2005: concept searching; March 23, 2006: concept search software applied to department’s work product; March 6, 2007: concept search software; April 4, 2006 #4 about concept-search software; and Feb. 19, 2006: concept search software.).

Cluster search, which groups terms and builds a search taxonomy (See my post of June 30, 2006: tools for online searching; Jan. 8, 2009: Clustify; and Aug. 26, 2008: mentions three search systems.).

We welcome comments

Your email address will not be published. Required fields are marked *