Rules ASCII Syntax: Ordered and Unordered Options

From: Harold Boley (boley@informatik.uni-kl.de)
Date: 05/27/03

  • Next message: Said Tabet: "DRAFT: List of Built-ins"
    Hi JC Colleagues,
    
    continuing our Datalog and Hornlog rules presentation at
    
    http://home.attbi.com/~stabet/ruleml/essential/essentialruleml.html
    
    as well as today's discussion about rules syntax, I have further developed
    the merger of Prolog and F-Logic syntax from "The Rule Markup Language":
    
    http://link.springer-ny.com/link/service/series/0558/bibs/2543/25430005.htm
    http://www.dfki.uni-kl.de/~boley/ruleml-mht.pdf
    
    Best,
    Harold
    
    
    
    Rules ASCII Syntax: Ordered and Unordered Options
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    
    
    Let us reconsider our Datalog example:
    
    discount(Customer,Product,"5.0 percent") :-
                                        premium(Customer), regular(Product).
    
    Note that this is a positional (ternary) discount relation, where you have
    to remember in queries that the first argument is the customer, the second
    is the product, and the third is the rebate obtained. This is in the spirit
    of XML's ordered children. Also note that the comma infix within n-ary
    relation applications for n>1 can be regarded as the order-imposing sequence
    operator (while for n=0 and n=1 the sequencing degenerates). 
    
    Reliance on order can be avoided in the spirit of RDF, DL and F-Logic by
    introducing properties, attributes, or roles. For example, an F-Logic-style
    version of the above rule can be written as follows:
    
    discount(cust->Customer;prod->Product;rebate->"5.0 percent") :-
                            premium(cust->Customer), regular(prod->Product).
    
    Note that in this object-centered ('object-oriented') modeling of relations
    the semicolon infix indicates an unordered set of n slots, each consisting
    of a 'role->filler' pair. Queries can now be posed without need for any
    order information (however, knowledge of the role names cust, prod, rebate
    is needed instead): Generalized (role) unification takes care for pairing
    up identical roles before recursively unifying their fillers (we don't go
    into the issue of implicit/explicit 'rest' variables here). The query
    
    discount(rebate->Amount;prod->"Honda";cust->"Peter Miller")
    
    thus succeeds if premium(cust->"Peter Miller") and regular(prod->"Honda")
    were stored as facts, binding Amount to "5.0 percent". 
    
    However, for unary relations like premium and regular, as well as for
    some binary relations, the role (->) syntax may become cumbersome.
    So in OO RuleML (http://www.ruleml.org/indoo) we permit rules using both
    ordered and object-oriented relations as premises as well as conclusions.
    In the example, with an object-oriented discount relation and (degenerately)
    ordered premium and regular relations, we obtain the following rule:
    
    discount(cust->Customer;prod->Product;rebate->"5.0 percent") :-
                                        premium(Customer), regular(Product).
    
    Finally, suppose we start again with positional Datalog rules, but then
    want to add optional spatio-temporal coordinates. Instead of extending
    the argument sequences of all affected relations by further ordered
    arguments, it may be preferable to add these extra bits in the unordered,
    object-oriented way. Restricting the rule scope to the region of MA and
    the period of 2003/04, our example thus becomes:
    
    discount(Customer,Product,"5.0 percent";region->"MA";period->"2003/04") :-
                      premium(Customer;period->"2003/04"), regular(Product).
    
    We assume that the 'ordered infix' (,) has precedence over the 'unordered
    infix' (;), so that the ordered sequence as a whole is connected with the
    "->" slots via a ";" (here, the one preceding the region role). The query
    
    discount("Peter Miller","Honda",Amount;period->When;region->Where)
    
    succeeds if premium("Peter Miller";period->"2003/04") and regular("Honda")
    were stored as facts, binding Amount to "5.0 percent", When to "2003/04",
    and Where to "MA".
    
    
    For Hornlog rules, complex terms are enriched by roles in the same fashion.
    
    In the example, the individual constant "5.0 percent" can be analyzed as the
    cterm percent[5.0], and the constant "2003/04" can be analyzed as either a
    positional cterm interval["2003","2004"] or again as an object-oriented cterm
    interval[start->"2003";end->"2004"]. Note that square brackets are preferable
    for cterms to clearly set them off from the parenthesized atoms.
    
    Consistently substituting these cterms for the constant occurrences in the
    above OO Datalog rule gives us one of the following OO Hornlog rules:
    
    discount(Customer,Product,percent[5.0];
             region->"MA";
             period->interval["2003","2004"])
     :-
        premium(Customer;period->interval["2003","2004"]),
        regular(Product).
    
    
    discount(Customer,Product,percent[5.0];
             region->"MA";
             period->interval[start->"2003";end->"2004"])
     :-
        premium(Customer;period->interval[start->"2003";end->"2004"]),
        regular(Product).
    
    
    These options are offered to accomodate various positional and object-oriented
    modeling styles. They can all be marked up in OO RuleML.
    


    This archive was generated by hypermail 2.1.4 : 05/27/03 EST