Open AccessGenre as Web search descriptor
Author(s) -
Rosso Mark A.
Publication year - 2005
Publication title -
proceedings of the american society for information science and technology
Language(s) - English
Resource type - Journals
eISSN - 1550-8390
pISSN - 0044-7870
DOI - 10.1002/meet.14504201291
Subject(s) - typology , relevance (law) , web page , computer science , information retrieval , context (archaeology) , set (abstract data type) , world wide web , identification (biology) , preference , history , botany , archaeology , biology , political science , microeconomics , law , economics , programming language
With the advent of large, digital collections of heterogenous documents (namely the World Wide Web, and large sub‐collections thereof), attention to the prospect of using document genre to constrain document search has increased. A major issue to be resolved in this endeavor is the identification of what document categories should be used as the genres. As genre is a “folk typology”, document categories must enjoy widespread recognition by their intended user groups, in order to qualify as genres. This poster briefly describes a series of user studies aimed at developing a set of web page categories that are widely recognized as genre. Results showed that, on average, over 70% of one study's 257 participants agreed on the genre of the 55 webpages shown to them, having chosen from a typology of 18 genres. In addition, to determine this genre typology's usefulness in the context of search, users were asked to judge the usefulness of genre‐annotated (vs. non‐annotated) webpage summaries in a series of simulated search tasks. Although results were inconclusive regarding the effect on the speed or accuracy of users' relevance judgements, many users reported a preference for the genre‐annotated webpage summaries when evaluating search results.