Hi Alvaro,
KIM does not do any kind of extraction of features on the document
level, except for the one you've mentioned. The way this is supposed to
work is to provide this kind of metadata additionally. There are two
ways to do that: by setting document level features to the kim
document; by providing a file with this metadata in XML and it will be
automatically assigned to the document. The second option works with
the population tool - part of the installation.
The place to find the format of this file is "your kim installation
path"/doc/sys-doc/kim-platform-administration/populate-doc-store.html
or through navigating the system documentation.
Somehow our focus never was on finding the document-level features,
just because KIM has been designed as a quite general platform for
semantic annotation, instead we have focused on finding entities in the
content.
On the other hand, a custom IE application can be created if you need
this functionality, especially if your input files are with a
predictable structure - this way you will be able to automatically
generate the features you would need for further processing.
all the best
borislav
Alvaro Hernandez wrote:
Thank you Borislav, but I have problems with this.
I tried to do it, but without success. From where does Kim get
this features (title, author, subject, etc..) ?, is it automatically ??
This is the output:
KIM Server connected.
Todo obtained successfully.
Document created from URL and annotated.
title null
[ Document's Features (begin) ]
[key: KeyEntities] [feature: ]
[key: MimeType] [feature: text/html]
[ Document's Features (end) ]
I attached my source. The document I used for the example has a
tag 'TITLE', does Kim get the title from this tag or from where??
Thank you a lot,
Dear Alvaro,
in the system documentation available in your installation you find
out about different ways of searching and accessing documents.
For the specific issue please look at the JavaDoc of the API and more
specifically into the KIMDocument interface documentation.
There you would find predefined constants for the most common feature
types on the document level.
The map of features itself is accessible through the getFeatures()
method.
all the best
borislav
Alvaro Hernandez wrote:
Any help, advice, tip???
Thank you.
Note: forwarded message attached.
Be a better Heartthrob. Get better relationship answers from
someone who knows.
Yahoo! Answers - Check it out.
Hi everybody,
I want to get features about a document, like title, author,
etc...,
Is there any way using the API ???
Thanks a lot,
Alvaro
Need a vacation? Get great deals to amazing places on
Yahoo! Travel.
_______________________________________________ NOTE: Please REPLY TO ALL to ensure that your reply reaches all members of this mailing list. KIM-discussion mailing list [email protected] http://ontotext.com/mailman/listinfo/kim-discussion_ontotext.com
No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.484 / Virus Database: 269.12.0/961 - Release Date: 8/19/2007 7:27 AM
Fussy? Opinionated? Impossible to please? Perfect. Join
Yahoo!'s user panel and lay it on us.
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.484 / Virus Database: 269.12.0/961 - Release Date: 8/19/2007 7:27 AM
|
_______________________________________________
NOTE: Please REPLY TO ALL to ensure that your reply reaches all members of this
mailing list.
KIM-discussion mailing list
[email protected]
http://ontotext.com/mailman/listinfo/kim-discussion_ontotext.com
