On 10 May 2012 04:28, Anand Chitipothu <[email protected]> wrote:
>
> On 09-May-2012, at 6:35 PM, Ben Companjen wrote:
>
>> Hi,
>>
>> Although I found 341 duplicates of President Clinton a lot yesterday,
>> there is still the "author" that goes by the name "Shirley
>> Institute/Wira Joint Conference (1977 Manchester, Eng.)". There are a
>> whopping 10,047 authors with that name! Merging those manually is only
>> for those who desperately need an extremely boring task :)
>>
>> Looking at the subject and book titles in the search results, I think
>> one MARC record was imported many times without duplicate detection,
>> so merging the authors would still leave some 10000 duplicate
>> works/editions.
>>
>> Any idea how to best solve this?
>
> I can work out an API. Will that help?
That is probably the best way to go.
It looks like the few authors that show up at the top of the search
results with multiple (= 2) works were incorrectly used as author of
those second works. I just corrected a few that according to the MARC
records should go under a Russian scientist and British Columbia's
Ministry of Finance.
I think there is only one author (the conference), and only one
edition of one work (The future of natural fibres), imported from one
MARC record [1] from the Miami University of Ohio ('Let's put the
item's price in, let's see, the 020 field. There, that looks nice." ;)
).
OL4602513A is the first key when ordering the keys in an ascending way.
Perhaps this can help determine what records (author, but also
work/edition) to merge or even delete?
Ben
[1]
http://openlibrary.org/show-records/marc_miami_univ_ohio/allbibs0016.out:7058864:1146
>
> Anand
> _______________________________________________
> Ol-discuss mailing list
> [email protected]
> http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss
> To unsubscribe from this mailing list, send email to
> [email protected]
_______________________________________________
Ol-discuss mailing list
[email protected]
http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss
To unsubscribe from this mailing list, send email to
[email protected]
