RECON – A Controlled English for BusinessRules

Ed Barkmeyer 1 Fabian Neuhaus 1,2

1National Institute of Standards and Technology 2Prometheus Computing

July 12th, 2013

What is RECON?

Dictionary & Vocabulary

Examples, examples, examples

Summary

How NIST became interested in Controlled English

I Requirements:I Represent domain experts knowledge about complex domainI Apply automatic reasoning

I Challenge: Find KR language that isI Semantically unambiguousI Highly expressiveI Easy to learn and use for domain experts

I Solution:compile→ logic language

How NIST became interested in Controlled English

I Requirements:I Represent domain experts knowledge about complex domainI Apply automatic reasoning

I Challenge: Find KR language that isI Semantically unambiguousI Highly expressiveI Easy to learn and use for domain experts

I Solution: restricted Englishcompile→ logic language

How NIST became interested in Controlled English

I Requirements:I Represent domain experts knowledge about complex domainI Apply automatic reasoning

I Challenge: Find KR language that isI Semantically unambiguousI Highly expressiveI Easy to learn and use for domain experts

I Solution: RECON languagecompile→ logic language IKL

Example: RECONcompile→ IKL

RECON Every person who attends RuleML is located in Seattle.

IKL ( f o r a l l (? p e r s o n 1 )( i f

( and( p e r s o n ? p e r s o n 1 )( p e r s o n . a t t e n d s . c o n f e r e n c e ? p e r s o n 1 RuleML )

)( p e r s o n . i s l o c a t e d i n . t h i n g ? p e r s o n 1 S e a t t l e )

) )

Example: RECONcompile→ IKL

RECON Every person who attends RuleML is located in Seattle.

IKL ( f o r a l l (? p e r s o n 1 )( i f

( and( p e r s o n ? p e r s o n 1 )( p e r s o n . a t t e n d s . c o n f e r e n c e ? p e r s o n 1 RuleML )

)( p e r s o n . i s l o c a t e d i n . t h i n g ? p e r s o n 1 S e a t t l e )

) )

The big picture

Features of Approach

I Based on English words and usageI Reads like EnglishI Supports domain vocabularies

I Formal grammarI strongly limits freedom of expressionI unique parseI unique translation to IKL

I Writing the language requires training

IKL

I Extension of ISO Common Logic by Pat Hayes & Chris Menzel

I (Syntactically) Higher-order logic

I Nominalized propositions (e.g. ‘that it rains’)

I Enables ‘modal’ expressions (e.g. ‘It is required that the fieldis watered’)

RECON vs. SBVR

I Both expressive languages

I RECON semantics = mapping to IKL ◦ IKL model theory

I SBVR has no formal semantics

Dictionary & Vocabulary

Reminder

Dictionary

I Dictionary consists of word forms

I No semantics

I Example for dictionary entryDictionary Verb: run runs ran running run

Vocabulary

I Vocabulary = collection of terminological entriesI Terminological entry = collection of declarations

I Primary term (mandatory)I Alternative formsI (Formal) definitions in RECON languageI Free text definitions / comments

I Terminological entry belongs to a syntactic category

Example Vocabulary

Name: Bride of Neptune

Type Noun: tanker

Mass Noun: gasoline

Adjective: (thing) is registered

Verb: (party) ships (shipment)

Alternative: (shipment) is shipped by (party)

Property: (party) is the supplier () for (shipment)

Unit: gallon: volume

Examples, examples, examples

Example: Simple sentence

Bride of Neptune is a registered tanker.

( e x i s t s (? t a n k e r 1 )( and

( and( t a n k e r ? t a n k e r 1 )( t h i n g . i s r e g i s t e r e d ? t a n k e r 1 ) )

(= B r i d e o f N e p t u n e ? t a n k e r 1 ) ) )

Example: Quantification

Every supplier ships some shipment.

( f o r a l l (? s u p p l i e r 1 )( i f

( s u p p l i e r ? s u p p l i e r 1 )( e x i s t s (? sh ipment2 )

( and( sh ipment ? sh ipment2 )( p a r t y . s h i p s . sh ipment ? s u p p l i e r 1 ? sh ipment2 )

) ) ) )

Examples: Connectives

Connectives are allowed both between sentences and noun phrases

I ACME is registered or ACME is not registered.

I ACME owns both Bride of Neptune and Titanic.

Example: Qualifiers

Any shipment that is shipped via Bride of Neptune is registered.

( f o r a l l (? sh ipment1 )( i f

( and( sh ipment ? sh ipment1 )( sh ipment . i s s h i p p e d v i a . v e s s e l

? sh ipment1 B r i d e o f N e p t u n e ) )( t h i n g . i s r e g i s t e r e d ? sh ipment1 ) ) )

Examples: Properties – dual nature

ACME is the supplier for SH12345. [verb]The supplier for SH12345 ships SH12345. [noun]

( f o r a l l (? t h i n g 1 )( i f

( and( t h i n g ? t h i n g 1 )( t h i n g . i s t h e s u p p l i e r f o r . sh ipment

? t h i n g 1 SH12345 ) )( p a r t y . s h i p s . sh ipment ? t h i n g 1 SH12345 ) ) )

Example: Measurements, quantities, and mass nouns

SH12345 consists of 1000 gallons of gasoline.

( e x i s t s (? g a s o l i n e 1 )( and

( and( g a s o l i n e ? g a s o l i n e 1 )( q u a n t i t y . i s t h e v o l u m e o f . t h i n g

( Qvalue 1000 " g a l l o n ") ? g a s o l i n e 1 ) )( sh ipment . c o n s i s t s o f . t h i n g SH12345 ? g a s o l i n e 1 ) ) )

Example: Deontic rules

Every shipment must be registered.

( o b l i g a t i o n ( t h a t( f o r a l l (? sh ipment1 )

( i f( sh ipment ? sh ipment1 )( t h i n g . i s r e g i s t e r e d ? sh ipment1 ) ) ) ) )

Example: Nominalized propositions

NIST prevents the situation where Ed is located in Seattle.

( f o r a l l (? s i t u a t i o n 1 )( i f

( and( s i t u a t i o n ? s i t u a t i o n 1 )( t h i n g . hasTheme . t h i n g ? s i t u a t i o n 1

( t h a t( p e r s o n . i s l o c a t e d i n . t h i n g Ed S e a t t l e )

) ) )( p e r s o n . p r e v e n t s . s i t u a t i o n NIST ? s i t u a t i o n 1 )

) )

Summary of interesting features

I n-ary verbsI Compatible with Davidsonian events

I Boolean connectives

I Quantifiers

I Properties

I Quantities and units of measurements

I Nominalized propositions

I Modals

I Collections

Thank you

https://sourceforge.net/projects/nistreconst/files/?source=navbar