Abstract
In this paper we describe how the Predictive Model Markup Language (PMML) standard enhances the JBoss Drools production rule engine with native support for using predictive models in business rules. The historic debate between symbolic and connectionist approaches to rule/model orchestration provides numerous examples of hybrid systems combining "hard" and "soft" computing techniques to achieve di-erent levels of integration. Rules are often used to decide when and which model to invoke; model outputs, in turn, can be used to evaluate the preconditions of a rule. In a loosely coupled system, the rule engine calls an external component implementing the predictive model, but this has several disadvantages, most notably the need to setup proper communications and reconcile any di-erence in the way the components encode the data. We propose instead, a tightly integrated system where predictive models and rules become part of the same reasoning framework. The models, encoded using the PMML 4 standard, are loaded and processed by a compiler implemented using the rule engine it-self. The PMML document is transformed into a set of facts that de-ne the model, and a series of rules that formalize the model's behavior. In addition, most PMML data processing, validation, and transformation procedures are also implemented using auto-generated rules. Finally, in oder to integrate model inputs and outputs seamlessly in the inference process, we exploit an extension of the Drools engine which adds native support for uncertainty and/or fuzziness.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings of the PMML Workshop 2011 - Held in Conjunction with the KDD-2011 Conference |
| Pages | 39-47 |
| Number of pages | 9 |
| DOIs | |
| State | Published - Sep 26 2011 |
| Externally published | Yes |
| Event | 2011 Workshop on Predictive Model Mark-up Language, PMML'11 - Held in Conjunction with the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD-2011 - San Diego, CA, United States Duration: Aug 21 2011 → Aug 21 2011 |
Other
| Other | 2011 Workshop on Predictive Model Mark-up Language, PMML'11 - Held in Conjunction with the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD-2011 |
|---|---|
| Country | United States |
| City | San Diego, CA |
| Period | 8/21/11 → 8/21/11 |
Fingerprint
Keywords
- PMML
- Predictive models
- Rule-based systems
ASJC Scopus subject areas
- Software
- Information Systems
Cite this
Enhancing a production rule engine with predictive models using PMML. / Sottara, Davide; Mello, P.; Sartori, C.; Fry, E.
Proceedings of the PMML Workshop 2011 - Held in Conjunction with the KDD-2011 Conference. 2011. p. 39-47.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Enhancing a production rule engine with predictive models using PMML
AU - Sottara, Davide
AU - Mello, P.
AU - Sartori, C.
AU - Fry, E.
PY - 2011/9/26
Y1 - 2011/9/26
N2 - In this paper we describe how the Predictive Model Markup Language (PMML) standard enhances the JBoss Drools production rule engine with native support for using predictive models in business rules. The historic debate between symbolic and connectionist approaches to rule/model orchestration provides numerous examples of hybrid systems combining "hard" and "soft" computing techniques to achieve di-erent levels of integration. Rules are often used to decide when and which model to invoke; model outputs, in turn, can be used to evaluate the preconditions of a rule. In a loosely coupled system, the rule engine calls an external component implementing the predictive model, but this has several disadvantages, most notably the need to setup proper communications and reconcile any di-erence in the way the components encode the data. We propose instead, a tightly integrated system where predictive models and rules become part of the same reasoning framework. The models, encoded using the PMML 4 standard, are loaded and processed by a compiler implemented using the rule engine it-self. The PMML document is transformed into a set of facts that de-ne the model, and a series of rules that formalize the model's behavior. In addition, most PMML data processing, validation, and transformation procedures are also implemented using auto-generated rules. Finally, in oder to integrate model inputs and outputs seamlessly in the inference process, we exploit an extension of the Drools engine which adds native support for uncertainty and/or fuzziness.
AB - In this paper we describe how the Predictive Model Markup Language (PMML) standard enhances the JBoss Drools production rule engine with native support for using predictive models in business rules. The historic debate between symbolic and connectionist approaches to rule/model orchestration provides numerous examples of hybrid systems combining "hard" and "soft" computing techniques to achieve di-erent levels of integration. Rules are often used to decide when and which model to invoke; model outputs, in turn, can be used to evaluate the preconditions of a rule. In a loosely coupled system, the rule engine calls an external component implementing the predictive model, but this has several disadvantages, most notably the need to setup proper communications and reconcile any di-erence in the way the components encode the data. We propose instead, a tightly integrated system where predictive models and rules become part of the same reasoning framework. The models, encoded using the PMML 4 standard, are loaded and processed by a compiler implemented using the rule engine it-self. The PMML document is transformed into a set of facts that de-ne the model, and a series of rules that formalize the model's behavior. In addition, most PMML data processing, validation, and transformation procedures are also implemented using auto-generated rules. Finally, in oder to integrate model inputs and outputs seamlessly in the inference process, we exploit an extension of the Drools engine which adds native support for uncertainty and/or fuzziness.
KW - PMML
KW - Predictive models
KW - Rule-based systems
UR - http://www.scopus.com/inward/record.url?scp=80053049517&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80053049517&partnerID=8YFLogxK
U2 - 10.1145/2023598.2023604
DO - 10.1145/2023598.2023604
M3 - Conference contribution
AN - SCOPUS:80053049517
SN - 9781450308373
SP - 39
EP - 47
BT - Proceedings of the PMML Workshop 2011 - Held in Conjunction with the KDD-2011 Conference
ER -