10:43 a.m.
Hi everybody,
I am using Drools as a part of my Bachelor Thesis.
My job is to build an expert system for medical diagnoses.
My Question:
Is there a best practice known in modeling a variable
list of questions and answers in Drools?
Up to now I modeled the list of questions implementing
Java Classes and Enums for each question/answer pair.
But to be able to change those options easyly I need
a higher abstraction level than simple Java Classes and Enums.
Could someone show me an example how to implement this?
I hope I have explained my question understandable :)
--
Dirk Conzelmann
IT-Services
Telefon: +49.179.2237995
Email: info(a)dirk-conzelmann.de
Web: www.dirk-conzelmann.de
11:46 a.m.
Hi Dirk,
Good to see that I am not the only one fighting with that question :-).
I guess the general answer is: it depends on how large your "enums"
might grow.
For instance if your application is eventually supposed to help
diagnosing (say) all MeSH diseases based on all MeSH symptoms, then an
approach where every symptom and disease is a Java class is not an
option; this would lead to so many classes that I doubt Drools user
interface will handle that, or will be useful anyway (drop-down lists
with thousands of items).
The opposite approach consists in creating a few "root" Java classes
like (say) Disease and Symptom and to store the MeSH hierarchy into a
"name" attribute.
The obvious drawback of the approach is that when your end-users will
have to write the rules, they will be left with problems like:
when
Symptom(name == "well... what's the name of this MeSH category??")
then
Disease(name == "hmmm... can't remember the exact name of the
disease as normalized in MeSH...")
end
I've posted a message on this topic (subject: Thoughts about rule
authoring) and Michael Anstis kindly suggested a technical answer. I am
afraid it did not fit my needs, but it could fit yours :-).
For the moment I have no solution (like an intermediate approach) to my
problem: I am stuck with approach #2.
I have the impression that my problem requires a very tight integration
between my database (MeSH) and the Drools suggestion engine.
If you have another approach, I would love to know it ;-).
Best regards,
Bruno.
Le 08/11/2011 17:43, Dirk Conzelmann a écrit :
Hi everybody,
I am using Drools as a part of my Bachelor Thesis.
My job is to build an expert system for medical diagnoses.
My Question:
Is there a best practice known in modeling a variable
list of questions and answers in Drools?
Up to now I modeled the list of questions implementing
Java Classes and Enums for each question/answer pair.
But to be able to change those options easyly I need
a higher abstraction level than simple Java Classes and Enums.
Could someone show me an example how to implement this?
I hope I have explained my question understandable :)
12:28 p.m.
You can use dynamic enums in Guvnor (that create dynamic populated
listboxes).
See guvnor's doc chapter "1.4.2.4.8. Data enumerations (drop down list
configurations)"
For your case where number of items are really huge, you may consider
using custom editors (again, in Guvnor), and implement a js page that
contains some listbox (as many as level of classification in your
hierarchy) or directly a tree to select the value. See "custom editor"
and "working set" part of Guvnor's doc.
If you are under Eclipse, well, Drools suggestion engine modification
may be the only way.
1:47 p.m.
I am in the same situation (and I suspect a lot of people with complex
domain models are), which is: how do I let the user easily navigate my
domain model while writing a rule.
There is support for providing a custom form within a popup when a fact
constraint needs to be specified. You can present it to the user whenever
a fact constraint needs to be supplied to the rule. Since you are
providing the contents of the custom form, you can represent your domain
however you wish to, navigate it perhaps using a tree, or even let the
user perform a mini search to get to the "MeSH category".
The custom form does not currently work with DSLs though, if you are
planning on using DSL sentences. But you can write your own DSL widget
that does (not too difficult, if you know GWT).
For my project, the biggest utility value of drools is the standalone
editor, the Guvnor rest api and the fact that it is opensource (so that I
can customize it for my very complex needs)
Thanks
G. Patel
From: Bruno Freudensprung <bruno.freudensprung(a)temis.com>
To: Rules Users List <rules-users(a)lists.jboss.org>
Date: 11/08/2011 10:49 AM
Subject: Re: [rules-users] Data Modeling for medical expert system
Sent by: rules-users-bounces(a)lists.jboss.org
Hi Dirk,
Good to see that I am not the only one fighting with that question :-).
I guess the general answer is: it depends on how large your "enums"
might grow.
For instance if your application is eventually supposed to help
diagnosing (say) all MeSH diseases based on all MeSH symptoms, then an
approach where every symptom and disease is a Java class is not an
option; this would lead to so many classes that I doubt Drools user
interface will handle that, or will be useful anyway (drop-down lists
with thousands of items).
The opposite approach consists in creating a few "root" Java classes
like (say) Disease and Symptom and to store the MeSH hierarchy into a
"name" attribute.
The obvious drawback of the approach is that when your end-users will
have to write the rules, they will be left with problems like:
when
Symptom(name == "well... what's the name of this MeSH category??")
then
Disease(name == "hmmm... can't remember the exact name of the
disease as normalized in MeSH...")
end
I've posted a message on this topic (subject: Thoughts about rule
authoring) and Michael Anstis kindly suggested a technical answer. I am
afraid it did not fit my needs, but it could fit yours :-).
For the moment I have no solution (like an intermediate approach) to my
problem: I am stuck with approach #2.
I have the impression that my problem requires a very tight integration
between my database (MeSH) and the Drools suggestion engine.
If you have another approach, I would love to know it ;-).
Best regards,
Bruno.
Le 08/11/2011 17:43, Dirk Conzelmann a écrit :
Hi everybody,
I am using Drools as a part of my Bachelor Thesis.
My job is to build an expert system for medical diagnoses.
My Question:
Is there a best practice known in modeling a variable
list of questions and answers in Drools?
Up to now I modeled the list of questions implementing
Java Classes and Enums for each question/answer pair.
But to be able to change those options easyly I need
a higher abstraction level than simple Java Classes and Enums.
Could someone show me an example how to implement this?
I hope I have explained my question understandable :)
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
-----------------------------------------
The information contained in this communication (including any
attachments hereto) is confidential and is intended solely for the
personal and confidential use of the individual or entity to whom
it is addressed. If the reader of this message is not the intended
recipient or an agent responsible for delivering it to the intended
recipient, you are hereby notified that you have received this
communication in error and that any review, dissemination, copying,
or unauthorized use of this information, or the taking of any
action in reliance on the contents of this information is strictly
prohibited. If you have received this communication in error,
please notify us immediately by e-mail, and delete the original
message. Thank you
2:23 a.m.
Hi,
Thanks a lot, to you and Vincent, for your very interesting answers!
So far I decided to use Drools without Guvnor (only Eclipse) but now I
realize I really should take a close look at it.
Best regards,
Bruno.
Le 08/11/2011 20:47, GPatel(a)tsys.com a écrit :
I am in the same situation (and I suspect a lot of people with
complex
domain models are), which is: how do I let the user easily navigate my
domain model while writing a rule.
There is support for providing a custom form within a popup when a
fact constraint needs to be specified. You can present it to the user
whenever a fact constraint needs to be supplied to the rule. Since you
are providing the contents of the custom form, you can represent your
domain however you wish to, navigate it perhaps using a tree, or even
let the user perform a mini search to get to the "MeSH category".
The custom form does not currently work with DSLs though, if you are
planning on using DSL sentences. But you can write your own DSL widget
that does (not too difficult, if you know GWT).
For my project, the biggest utility value of drools is the standalone
editor, the Guvnor rest api and the fact that it is opensource (so
that I can customize it for my very complex needs)
Thanks
G. Patel
4:16 a.m.
Hi Bruno,
On Tue, 08 Nov 2011 18:46:45 +0100, Bruno Freudensprung wrote:
Hi Dirk,
Good to see that I am not the only one fighting with that question
:-).
I guess the general answer is: it depends on how large your "enums"
might grow.
For instance if your application is eventually supposed to help
diagnosing (say) all MeSH diseases based on all MeSH symptoms, then
an
approach where every symptom and disease is a Java class is not an
option; this would lead to so many classes that I doubt Drools user
interface will handle that, or will be useful anyway (drop-down lists
with thousands of items).
Yes, Exactly!
The opposite approach consists in creating a few "root" Java classes
like (say) Disease and Symptom and to store the MeSH hierarchy into a
"name" attribute.
The obvious drawback of the approach is that when your end-users will
have to write the rules, they will be left with problems like:
when
Symptom(name == "well... what's the name of this MeSH
category??")
then
Disease(name == "hmmm... can't remember the exact name of the
disease as normalized in MeSH...")
end
hehe yes, yes :)
I've posted a message on this topic (subject: Thoughts about rule
authoring) and Michael Anstis kindly suggested a technical answer. I
am
afraid it did not fit my needs, but it could fit yours :-).
Ok, i'll search
for that later.
For the moment I have no solution (like an intermediate approach) to
my
problem: I am stuck with approach #2.
I have the impression that my problem requires a very tight
integration
between my database (MeSH) and the Drools suggestion engine.
If you have another approach, I would love to know it ;-).
Yes, i'll let you
know. btw. i received a pretty interesting message
from 'davide'
a few moments ago.
http://www.jboss.org/tohu
https://github.com/droolsjbpm/drools-chance/tree/master/drools-informer
Maybe this will help you too :)
Best regards,
Bruno.
Le 08/11/2011 17:43, Dirk Conzelmann a écrit :
> Hi everybody,
>
> I am using Drools as a part of my Bachelor Thesis.
> My job is to build an expert system for medical diagnoses.
>
> My Question:
> Is there a best practice known in modeling a variable
> list of questions and answers in Drools?
>
>
> Up to now I modeled the list of questions implementing
> Java Classes and Enums for each question/answer pair.
>
> But to be able to change those options easyly I need
> a higher abstraction level than simple Java Classes and Enums.
>
>
> Could someone show me an example how to implement this?
>
>
> I hope I have explained my question understandable :)
>
>
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
--
Dirk Conzelmann
IT-Services
Telefon: +49.179.2237995
Email: info(a)dirk-conzelmann.de
Web: www.dirk-conzelmann.de
2:53 a.m.
Hi Dirk,
you might want to take a look at this:
http://www.jboss.org/tohu
This is an interesting but rather old project. We're also working on a
slightly refactored version here:
https://github.com/droolsjbpm/drools-chance/tree/master/drools-informer
The project provides a dynamic "question/answer" framework and supports GUI
interfaces.
As for the /content/ of the questions, it seems that you need some medical
vocabulary.
Hand-coding enum-equivalent lists may be an overwhelming task, depending on
how "realistic"
your system has to be. We're dealing with similar problems in another
project, including the
predictive modelling techniques for the actual inference.
If you want to share thoughts, let me know.
Cheers
Davide
--
View this message in context:
http://drools.46999.n3.nabble.com/rules-users-Data-Modeling-for-medical-e...
Sent from the Drools: User forum mailing list archive at Nabble.com.
4:11 a.m.
Hi Davide and the other users too,
thank you all for your interesting links. I´ll think
that 'tohu' really seams the solution for that problem.
I think it is rather interesting that after spending
my first weeks on Drools reading the documentation
I never saw any link pointing to 'tohu'. :)
Thanks very much, I´ll share my experiences.
Despite from being interested in this aproach,
I have to adjust my tasks for being able
to finish my Bachelor Thesis in the given amount of time.
On Wed, 9 Nov 2011 00:53:18 -0800 (PST), Davide Sottara wrote:
Hi Dirk,
you might want to take a look at this:
http://www.jboss.org/tohu
This is an interesting but rather old project. We're also working on
a
slightly refactored version here:
https://github.com/droolsjbpm/drools-chance/tree/master/drools-informer
The project provides a dynamic "question/answer" framework and
supports GUI
interfaces.
As for the /content/ of the questions, it seems that you need some
medical
vocabulary.
Hand-coding enum-equivalent lists may be an overwhelming task,
depending on
how "realistic"
your system has to be. We're dealing with similar problems in another
project, including the
predictive modelling techniques for the actual inference.
If you want to share thoughts, let me know.
Cheers
Davide
--
View this message in context:
http://drools.46999.n3.nabble.com/rules-users-Data-Modeling-for-medical-e...
Sent from the Drools: User forum mailing list archive at Nabble.com.
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
--
Dirk Conzelmann
IT-Services
Telefon: +49.179.2237995
Email: info(a)dirk-conzelmann.de
Web: www.dirk-conzelmann.de
12:38 p.m.
Tohu is a relatively old, unofficial and inactive project. We're refactoring
that as "drools-informer",
but that's work in progress and I'm not sure the jQuery web client that
comes with tohu would still work. So I'd recommend to stick to tohu for your
thesis (especially giving your time constraints :))
Feel free to contact me (dsotty(a)gmail.com) if you have questions. I'm
sort-of-in-co-charge of monitoring drools research and I normally tutor
bachelor/master students at the university. So, I'd be happy to share advice
in exchange for information on where and how drools is being used as a
research platform ;)
--
View this message in context:
http://drools.46999.n3.nabble.com/rules-users-Data-Modeling-for-medical-e...
Sent from the Drools: User forum mailing list archive at Nabble.com.
4:36 a.m.
Hi Davide,
thank you for your advices. I'll give it a shot and
will share my experiences.
Sincerly
On Wed, 9 Nov 2011 10:38:35 -0800 (PST), Davide Sottara wrote:
Tohu is a relatively old, unofficial and inactive project. We're
refactoring
that as "drools-informer",
but that's work in progress and I'm not sure the jQuery web client
that
comes with tohu would still work. So I'd recommend to stick to tohu
for your
thesis (especially giving your time constraints :))
Feel free to contact me (dsotty(a)gmail.com) if you have questions. I'm
sort-of-in-co-charge of monitoring drools research and I normally
tutor
bachelor/master students at the university. So, I'd be happy to share
advice
in exchange for information on where and how drools is being used as
a
research platform ;)
--
View this message in context:
http://drools.46999.n3.nabble.com/rules-users-Data-Modeling-for-medical-e...
Sent from the Drools: User forum mailing list archive at Nabble.com.
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
--
Dirk Conzelmann
IT-Services
Telefon: +49.179.2237995
Email: info(a)dirk-conzelmann.de
Web: www.dirk-conzelmann.de
4813
days inactive
4815
days old
9 comments
5 participants
participants (5)
-
Bruno Freudensprung -
Davide Sottara -
Dirk Conzelmann -
GPatel＠tsys.com -
Vincent Legendre