This is an archive of an inactive wiki and cannot be modified.

This is one of the possible Use Cases.

Abstract

This use case shows the need of uncertainty representation in personalisation. For example, an online service can help its clients to find products (such as books) with reasonable price. Note that different clients might have different "reasonable prices". The following user-defined function

calculates the values of "resonable" for the price ?x, given the reference maximum price ?rp. For instance, reasonable_price(30,50)=0.4, while reasonable_price(80,50)=0.

This use case provides a scenario where datatype built-ins can be used to represent some levels of uncertainty in rules. The application aims at personalisations. It requires reasoning with rules and datatype built-ins, such as those defined in XQuery and XPath.

Status

Links to Related Use Cases

This use case is related to the work in the Fuzzy RuleML technical group (http://image.ntua.gr/FuzzyRuleML/).

Relationship to OWL/RDF Compatibility

The semantics of datatype built-ins should be compatibale with RDF and OWL datatypes.

Requirements on the RIF

Narratives

An online servise provides suggestions on hotels reasonably near business locations and with reasonable price. Its suggestions are expected to be based on clients' personal preference, i.e., the price that she finds reasonable and the distance that she thinks is accpetable.

1) Personalised closeness can be defined by the following user-defined function:

2) Personalised reasonable prices can be defined by the following user-defined function:

3) Candidate hotels w.r.t some business location for a clients can be decided by the following rule

Commentary

In the above example, closeness and reasonable prices are vague concepts, and are represented by datatype-related functions.

[1] P. Vojtas. Fuzzy logic programming. Fuzzy Sets and Systems, 124:361-370, 2001.