Examples
Grandparent
Great-Grandparent
Ancestor
Metagol Web Interface
Metagol is an inductive logic programming (ILP) system based on the meta-interpretive learning framework.
Download on Github.
Predicates to use in background knowledge
Define the predicates that you would like to use in your background knowledge. You may define predicates with either one or two arguments.
Name
Number of arguments
add new
Arguments to use in background knowledge
Define the arguments that you would like to use in your background knowledge.
Name
add new
Edit background knowledge
Enter your background knowledge here. Each row consists of a predicate and its one or two arguments selected from those you've defined above.
Predicate
Argument 1
Argument 2
add row
Add metarules
Select the metarules you would like to use.
Metarule
add row
Learn a predicate
delete
Name of predicate to learn
This is the target predicate you are trying to learn.
Edit positive examples
These are positive examples of the target predicate (examples that are true), that will be used to learn the target predicate.
Predicate
Argument 1
Argument 2
add row
Edit negative examples
These are negative examples of the target predicate (examples that are false), that will be used to learn the target predicate.
Predicate
Argument 1
Argument 2
add row
Predicates to learn
Define the target predicates that you would like to learn, and the examples you will provide to learn from based on the background knowledge you've defined above. You can learn one or more predicates, in order.
Predicates
add predicate to learn
Once you have filled in your background knowledge and predicate to learn, generate the code that will run metagol by clicking this button. You can copy this code and save it to a .pl file if you want to run metagol locally on your system.
Code to execute:
Click this button to execute the code on the server, and see the results below.
Output