Information Architecture for the World Wide Web
by Louis Rosenfeld and Peter Morville


The unconfirmed error reports are from readers.  They have not yet been 
approved or disproved by the author or editor and represent solely the
opinion of the reader.

This page was updated April 11, 2008.

Here's a key to the markup:
        [page-number]: serious technical mistake
        {page-number}: minor technical mistake
        <page-number>: important language/formatting problem
        (page-number): language change or minor formatting problem
        ?page-number?: reader question or request for clarification


UNCONFIRMED errors and comments from readers: 


{142} footnote:
The SiteMap tool you mention on this page (http://www.jazzsoft.com) doesn't 
seem to exist anymore.

According to Network Solutions, the domain is still registered (to Keibock 
Lee), but the server will not respond.


[159] 2nd paragraph- 'Recall and Precision';
This section provides formalas for two key information retrieval priciples, recall and precision. The formula for precision is incorrect. Precision of search results is the number of relevant documents retreived by the query out of all the documents retrieved by the query. In other words, of all the documents my search retreived, how many are actually relevant. Hence, the denominator of the formula provided in te book needs correction.

Precision= # relevant documents retrieved/# total documents retreived

Testing for how many relevant docs were retrieved out of all the docs in the collection (as your current precision formula implies) is not very useful since it doesn't tell you whether more relevant docs were there to be had or not. Example, if get 2 relevant docs out of the whole collection, maybe there were only two, maybe there were 8 more. Who knows? However, if only 2 out of 10 docs retrieval are relevant (by the correct Precision formula), that tells me something about the accuracy or "precision" of my system's matching capability.

Also, if I know that there are 10 relevant docs in the collection and I get 8 relevant docs in my results (by the Recall formula), that tells me something about the breadth of my system's matching capability.