Re: Leveraging Authority Data in Keyword Searches

But the departmental authority records are _linked_ to the parent 
authority record, right?

You just need your indexer to take that into account and follow the link 
and include those terms in the indexed fields for the "child" records.

Getting this to actually work right and give the users the most 
'relevant' response for a query is going to be tricky. But the info is 
there.

Doesn't mean it's going to be easy to take advantage of it in a 
vendor-supplied ILS. It won't be.

Jonathan

Ed Jones wrote:
> Unfortunately, authority records aren't currently designed to support keyword searching.  They were designed in a world of card catalogs, which could only be approached via browsing, so their designed hierarchically.  Consequently, a keyword search such as "university munich mathematics" (which works beautifully in Google) fails utterly in library catalogs because "university of munich" occurs only in a reference on the authority record for "Universität München", not in references on any of the authority records for its academic departments.  Similarly, variant personal names do not occur on authority records for an author's works.  One way to get around this--at least to some degree--might be to include all the synonymous strings (e.g., "Department of Mathematics at the University of Munich") that occur in 670 "Sources Found" notes on a given authority record as explicit 4XX references.  This would increase the probability of a match between what's in a given user's m!
 in!
>  d and what's in a given authority record.
>
> Ed Jones
> National University (San Diego, Calif.)
>
> -----Original Message-----
> From: Next generation catalogs for libraries [mailto:NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of Jacobs, Jane W
> Sent: Friday, April 24, 2009 11:31 AM
> To: NGC4LIB_at_LISTSERV.ND.EDU
> Subject: [NGC4LIB] Leveraging Authority Data in Keyword Searches
>
> I don't know if this has been discussed before, but I'm interested in
> leveraging our authority records for keyword searches in our
> "traditional" ILS.  
>
>  
>
> For example a search using "author keywords" or "anywhere keywords" for
> "Gafenco, Gregorio" gets no results, yet we have works for: Gafencu,
> Grigore, 1892- 
>
> With a cross reference from: Gafenco, Gregorio, 1892-
>
>  
>
> In my mind, this ought not to be rocket science, but I'm not quite sure
> how the algorithm would work.  Something like:
>
>  
>
> Basic Keyword search in Bibliographic records +
>
>  
>
> Keyword search in Type z (authority) records fields 1XX & 4XX -> 
>
> if a hit occurs grab the 1XX and perform appropriate keyword search
> (i.e. author keyword, title keyword, etc.) on it and add to result set.
>
>  
>
> Does that make sense?  Would you get too many irrelevant hits?  Would
> this kill your processors?
>
> Is anyone doing it?  If so which system?
>
>  
>
> Thanks in advance for anyone's thoughts!
>
> JJ
>
>  
>
>  
>
> **Views expressed by the author do not necessarily represent those of
> the Queens Library.**
>
>  
>
> Jane Jacobs
>
> Asst. Coord., Catalog Division
>
> Queens Borough Public Library
>
> 89-11 Merrick Blvd.
>
> Jamaica, NY 11432
>
> tel.: (718) 990-0804
>
> e-mail: Jane.W.Jacobs_at_queenslibrary.org
>
> FAX. (718) 990-8566
>