search

carneades / carneades-4

Version 4 of Carneades, implemented in Go
https://carneades.github.io/
Other
48 stars 13 forks source link

Defining issues for compound terms #36

Closed pallix closed 7 years ago

pallix commented 7 years ago

Is it possible to somehow define issues for compound terms without knowing in advance what the compound term will be? For example in the example below the domain of person is infinite (the domain of the first argument is infinite, there is potentially an infinite numbers of persons) but for a given person p we would really want to have:

issues:
  i1: 
    positions: 
      - person(p, alive)
      - ¬person(p, alive)

and potentially also

issues:
  i2: 
    positions: 
      - person(p, alive)
      - person(p, dead)

Whenever a new person is expressed in the system (for example with an assumption such as born(lea)), we need to add issues for lea such as the one above.

Is there a trick to simplify that?

meta:
  title: Identifying an inconsistency on a proposition

language:
  inconsistent/2: "%s has an inconsistent instance %s"
  breathing/1: "%s is breathing"
  killed/1: "%s was killed"
  person/2: "person %s is %s"
  alive/0: "alive"

statements:
  person(p,alive): p is alive
  ¬person(p,alive): ¬p is alive

argument_schemes:

  - id: breathing
    conclusions:
      - person(p,alive)
    premises:
      - breathing(p)

  - id: killed
    conclusions:
      - ¬person(p,alive)
    premises:
      - killed(p)

assumptions:
  - breathing(p)
  - killed(p)
tfgordon commented 7 years ago

Yes. Carneades also has issue schemes, analogous to argumentation schemes. They are instantiated automatically to create issues. There should be some examples in the AGs/YAML directory.

On Thu, Mar 30, 2017 at 4:05 PM +0200, "Pierre Allix" notifications@github.com<mailto:notifications@github.com> wrote:

Is it possible to somehow define issues for compound terms without knowing in advance what the compound term will be? For example in the example below the domain of person is infinite (the domain of the first argument is infinite, there is potentially an infinite numbers of persons) but for a given person p we would really want to have:

issues: i1: positions:

and potentially also

issues: i2: positions:

Whenever a new person is expressed in the system (for example with an assumption such as born(lea)), we need to add issues for lea such as the one above.

Is there a trick to simplify that?

meta: title: Identifying an inconsistency on a proposition

language: inconsistent/2: "%s has an inconsistent instance %s" breathing/1: "%s is breathing" killed/1: "%s was killed" person/2: "person %s is %s" alive/0: "alive"

statements: person(p,alive): p is alive ¬person(p,alive): ¬p is alive

argument_schemes:

assumptions:

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/carneades/carneades-4/issues/36, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AAsP-1ZS-18YkYODen2fkdpLhukC8tDRks5rq7ahgaJpZM4MuaIw.

pallix commented 7 years ago

Ok I found them, thank you. But I cannot find how to specify the proof standard for a given issue_schemes.

tfgordon commented 7 years ago

Good point. It is not currently possible to set the proof standard in issue schemes. All issues generated are assigned the preponderance of evidence standard.

This was an oversight. I'll post an issue to fix this.

On Thu, Mar 30, 2017 at 8:29 PM +0200, "Pierre Allix" notifications@github.com<mailto:notifications@github.com> wrote:

Ok I found them, thank you. But I cannot find how to specify the proof standard for a given issue_schemes.

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/carneades/carneades-4/issues/36#issuecomment-290502518, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AAsP-5-XZoMfzkI_PuXE1x_md5AfV5agks5rq_R2gaJpZM4MuaIw.