WO2016068691A1

WO2016068691A1 - A system and method for ontology model process driven decision support system

Info

Publication number: WO2016068691A1
Application number: PCT/MY2015/050133
Authority: WO
Inventors: Yew Choong CHEW; Thiam Ho SECK; May Fern KOH; Rafizah RAHMAT; Nagendran M. Perumal
Original assignee: Mimos Berhad
Priority date: 2014-10-28
Filing date: 2015-10-16
Publication date: 2016-05-06
Also published as: MY187134A

Abstract

The system (100) and method (400) of the present invention provides a decision support system (DSS) 5 which utilizes open standards and information reuse that provides reusable solution for other open standard ontology system. The system's decision making process decouples the sub-systems in the DSS, processes, and data needed for deployment of a new decision making process, algorithms and supporting data without any modification to the system. The distinctiveness of the present invention lies in the following sub systems of the ontology model of the Decision Support System (DSS); Workflow Manager (104) for managing the workflow of a predefined solution; Data Integrator (110) for interacting with a variety of data from internal and external sources; Rule-based Inference Engine (106) for supporting automatic decision making based on pre-defined rules that interoperate with a range of rule-based reasoners in plug-and-play manner; and Algorithm Library (108) for processing customized algorithm to enable deployment of algorithms based on generic methodology defined in the Library without any modification of the ontology; algorithms are shared between heterogeneous implementers wherein each implementer is identified through a unique identification (ID) which is defined in the system; and a lookup directory is used to load and execute the algorithms in the Library.

Description

A SYSTEM AND METHOD FOR ONTOLOGY MODEL PROCESS DRIVEN DECISION SUPPORT SYSTEM FIELD OF INVENTION

The present invention relates to a system and method for ontology model process driven decision support system which utilizes open standards and information reuse that provides reusable solution for other open standard ontology model decision support system. In particular, the invention relates to systems and methods which decouple the sub systems in the decision support system that allows new algorithms, processes and supporting data required for deployment of a new decision making process to be deployed without any modification to the core system.

BACKGROUND ART

Existing Decision Support Systems (DSSs) does not encourage information sharing nor reuse out of the application domain. Further, different applications propose different taxonomies, classifications, algorithms and processes as each type of application depends on a specific way of processing data and making inference. There is no universally accepted taxonomy of different DSS. Data and process are usually two different types of entities which are usually tightly coupled in system code level. Typical DSS requires modification in its system code when there is a new decision making process. Migration from one DSS to another DSS requires a lot of reworks on models, algorithms and processes which is expensive and redundant. Ontology-based solution requires a Knowledge Base (KB) which requires continuous effort in transforming and updating the KB from the existing database(s) which is also expensive and redundant.

United States Patent Publication No. 201 1/191270 A1 (US '270 Publication) entitled: Intelligent Decision Supporting System and Method for Making Intelligent Decision relates to a technical field of intelligent decision, particularly, an intelligent decision supporting system and a method for making intelligent decision. The invention as disclosed in the US '270 Publication includes a multi-dimensional classifier that defines different semantic standards. Further, the system as disclosed in the US '270 Publication is trained based on different semantic standards for classifying text by the semantic standards and two separate modules for question and answer are provided in US '270 Publication (i.e. the Question Submitting Module for forming questions; and Decision Reply Module for providing answers to the questions).

In a paper entitled "A framework for ontology-based decision support system for e-learning modules, business modelling and manufacturing systems" by Arnab Bhattacharya; M.K.Tiwari; J. A. Harding; published in the Journal of Intelligent Manufacturing; October 2012, Volume 23, Issue 5, pp 1763-1781 ; Springer Link (Bhattacharya et. al), an ontology-based decision support is provided for e-learning modules, business modelling and manufacturing systems. In Bhattacharya et. al., a framework for ontology is provided by representing information in structured models by identifying how ontology models is used to define semantics and relationships in representing objects and modules.

In an IEEE paper entitled: Study on Ontology-Based Integration Strategy and Methods for PMS; Sheng Lu, Zhongjian Cai, Tan Liu; 2008 IEEE (IEEE paper), an ontology-based integration strategy which performs an ontology-based Decision Support System is provided. In the IEEE paper, an ontology-based Decision Support System comprising Natural Language Processing Agent; Fuzzy Inference Agent; and Performance Decision Support Agent. CMMI Ontology and project personal ontology that are predefined by the domain experts and project domain experts is utilized.

SUMMARY OF INVENTION

The present invention relates to a system and method for ontology model process driven decision support system which utilizes open standards and information reuse that provides reusable solution for other open standard ontology model decision support system. In particular, the invention relates to systems and methods which provide a generic platform that builds customized solution for any domains.

One aspect of the present invention provides a system (100) for ontology-based decision support which utilizes open standards and information reuse. The system comprising at least one Workflow Manager (104) for managing workflow of predefined solution; at least one Data Integrator (1 10) for interacting with variety of information from a plurality of sources; at least one Rule-based Inference Engine (106) for supporting automatic decision making based on pre-defined rules that interoperate with a range of rule-based reasoners; and at least one Algorithm Library (108) for processing algorithm to enable deployment of algorithms based on methodology defined in said Library without any modification of the ontology; algorithms are shared between heterogeneous implementers wherein each implementer is identified through a unique identification (ID) which is defined in the system.

Another aspect of the invention provides that the rule-based reasoners are in plug and play manner.

A further aspect of the invention provides that at least one Algorithm Library and the at least one Algorithm Library further comprising a lookup directory, said lookup directory is used to load and execute algorithms in said Library.

Yet another aspect of the invention provides that the said system decouples the at least one Workflow Manager, the at least one Data Integrator, the at least one Rule-based Inference Engine and the at least one Algorithm Library in said system together with information required for deployment of decision support.

Another aspect of the invention provides a method for (400) ontology-based decision support which utilizes open standards and information reuse. The method comprising steps of retrieving information from heterogeneous sources (402); predefining a set of semantic rules (404); mapping data values among semantic rules (406); configuring workflow and processes in ontology (408); and integrating data through ontology (410).

A further aspect of the invention provides that the step of configuring workflow and processes in ontology further comprises steps of accepting request (702) and invoke requested workflow of a predefined solution (704); generating ontology model (706); identifying process (708) and thereafter executing process (710); updating ontology model (712); and determining if process ended (714). If process is not ended, reiterating step 708 for next process; if process has end; retrieving ontology model (716); constructing (718) and returning response (720).

Yet another aspect of the invention provides that the step of attaching algorithms to processes further comprises steps of receiving request and trigger from Workflow Manager (802); identifying algorithm (804) and invoking said algorithm (806); accessing and updating ontology (808); retrieving ontology model (810); and returning to Workflow Manager (812).

Still another aspect of the invention provides that the step of creating rules for inference engine further comprises steps of receiving request and trigger from Workflow Manager (902); identifying rule (904) and invoking said rule (906); accessing and updating ontology (908); retrieving ontology model (910); and returning to Workflow Manager (912).

A further aspect of the invention provides that the step of integrating data through ontology further comprises steps of receiving request and trigger from Workflow Manager (1002); resolving data integration type (1004); determining selection of data integration type (1006); if data integration is of output type: accessing ontology model (1010); writing ontology to external data source (1012); and returning to Workflow Manager (1014); if data integration is of input type: mapping and creating external data source to ontology (1008).

The present invention consists of features and a combination of parts hereinafter fully described and illustrated in the accompanying drawings, it being understood that various changes in the details may be made without departing from the scope of the invention or sacrificing any of the advantages of the present invention.

BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS

To further clarify various aspects of some embodiments of the present invention, a more particular description of the invention will be rendered by references to specific embodiments thereof, which are illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail through the accompanying drawings in which:

FIG. 1.0 illustrates the general architecture of the system of the present invention.

FIG. 2.0 illustrates the primary ontology of the present invention.

FIG. 3.0 illustrates the primary ontology of Workflow Manager of the present invention.

FIG. 4.0 is a flowchart illustrating the methodology of the present invention.

FIG. 5.0 illustrates the model processes and the process chain of the present invention in which ontology is used to model a workflow.

FIG. 6.0 illustrates the process chain of Workflow Manager on the interconnection of the chain processes.

FIG. 7.0 is a flowchart illustrating the steps for configuring workflow and processes in ontology in the Workflow Manager.

FIG. 8.0 is a flowchart illustrating the steps for attaching algorithms to processes on the Algorithm Library. FIG. 9.0 is a flowchart illustrating the steps for creating rules for inference engine in the Rule-Based Inference Engine.

FIG. 10.0 is a flowchart illustrating the steps for integrating data through ontology in the Data Integrator. Table 1.0 defines the class annotations of the Decision Support System. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention provides a decision support system (DSS) that support and aid decision making. The system is a generic platform which utilizes open standards and information reuse that is used to build customized solution for any domains. In particular, the system's decision making process is driven by ontology and its supporting data is model in ontology to describe the available functionalities for each of its sub-systems. The functionalities required from each sub-system and its interconnection is described through the ontology process wherein the system decouples the sub-systems, processes, and data needed for decision making. Upon deployment of the system, a new decision making process, algorithms and supporting data is deployed separately without any modification to the system. Hereinafter, this specification will describe the present invention according to the preferred embodiments. It is to be understood that limiting the description to the preferred embodiments of the invention is merely to facilitate discussion of the present invention and it is envisioned without departing from the scope of the appended claims.

FIG. 1.0 illustrates a general architecture of the system of the present invention. As illustrated in Figure 1.0, the ontology model process-driven Decision Support System (100) comprising a Workflow Manager (104), Rule- based Inference Engine (106), Algorithm Library (108) and Data Integrator (1 10). The Workflow Manager (104) manages the workflow of a predefined solution which contains one or more processes. A generic method is provided wherein the workflow is required at the start process to invoke the required processes from the beginning to the end without any human interaction. When a request is received by the system from an external application, the Workflow Manager (104) calls a workflow defined in the deployed ontology and subsequently calls the other required processes in the process chain.

Data Integrator (1 10) interacts with a variety of information and data which includes data from a plurality of sources both internal and external sources. Data Integrator (1 10) also supports transformation of data between various formats (structured, semi-structured or unstructured). Further, Data Integrator (1 10) allows pre-defined data mapping between relational database and graph database as Data Integrator (1 10) can dynamically transform data while managing on-demand Knowledge Base based on real-time demand.

The Rule-Based Inference Engine (106) of the present invention supports automatic decision making based on pre-defined rules as the Rule-based Inference Engine (106) is designed to be generic to interoperate with a range of rule-based reasoners in a plug-and-play manner.

The Algorithm Library (108) of the present invention is an Algorithm Library which processes customized algorithm to aid the decision making process. A generic methodology is defined in the Algorithm Library (108) to enable deployment of algorithms without any modification of the ontology. The algorithms are shared between heterogeneous implementers. Each implementer is identified through a unique identification (ID) which is defined in the system (100). A lookup directory will be used to load and execute the algorithms in the Algorithm Library.

The system (100) decouples the Workflow Manager (104), the at least one Data Integrator (1 10), the at least one Rule-based Inference Engine (106) and the at least one Algorithm Library (108) in said system together with information required for deployment of decision support.

FIG. 2.0 illustrates the primary ontology of the present invention while Table 1.0 defines the class annotations of the Decision Support System (DSS) of the present invention. As illustrated in Figure 2.0, DSS is the core of the invention. In the DSS, a class can have subclass or subclasses that represent concepts that are more specific than the superclass. Dl (Data Integrator), RBIE (Rule-Based Inference Engine), AL (Algorithm Library) and WM (Workflow Manager) are specific subclasses of the DSS class wherein each of them represents a sub-system of the invention.

Referring to FIG. 4.0, a flowchart on the steps of the methodology of the present invention is illustrated. As illustrated in FIG. 4.0, the method (400) for ontology-based decision support which utilizes open standards and information reuse is initiated by retrieving information from heterogenous sources (402). Thereafter, a set of semantic rules are predefined (404) and data values are mapped among the semantic rules (406). Subsequently, workflow and processes are configured in ontology (408) wherein algorithms are attached to the processes. Rules for inference engine are created and data is integrated through ontology (410).

FIG. 3.0 illustrates the primary ontology of Workflow Manager (104) of the present invention while FIG. 5.0 illustrates the model processes and the process chain of the present invention in which ontology is used to model a workflow. FIG. 6.0 illustrates the process chain of the workflow on the interconnection of the chain processes. Referring to FIG. 3.0, the ontology of the Workflow Manager contains three subclasses (i.e. the process Dl, process RBIE and the process AL). Process Dl subclass contains all needed processes for data integration activities; process RBIE subclass contains all rules for inference; and process AL subclass contains all needed algorithm(s) for decision making process. Referring to FIG. 5.0, the process chain in ontology as illustrated in Figure 5.0 can be configured by the user from process Dl, process RBIE and the process AL of the ontology of the Workflow Manager. Ontology contains properties for describing its attributes and determining its condition. The WF class consists of a property called hasStartProcess, which indicates the starting process of the workflow. Each of the incurred process has a property called hasNextProcess, which indicate the next process to be execute and a hasNextProcess property with NIL value indicates the end of the workflow.

FIG. 7.0 is a flowchart illustrating the steps for configuring workflow and processes in ontology in the Workflow Manager. As illustrated in FIG. 7.0, the requester will trigger the DSS pre-defined workflow and all the processes in the sub-class stored in the workflow will be processed sequentially by initially accepting request (702) and invoking requested workflow of a predefined solution (704). Thereafter, ontology model is generated (706) and process is identified (708) and process is executed (710). Ontology model is updated (712) and it is further determined if process ended (714). If process is not ended, the process from step 708 is reiterated. If the process has end; ontology model is retrieved (716); response is constructed (718) and the results will be returned (720) to the requester upon completion of all of the process(s). FIG. 8.0 is a flowchart illustrating the steps for attaching algorithms to processes on the Algorithm Library. As illustrated in FIG. 8.0, the specified algorithm will be invoked receiving request and trigger from Workflow Manager (802). The algorithm is identified (804) and said algorithm is invoked (806). The result is stored in Ontology as ontology is accessed and updated (808). Thereafter, the result is obtained by retrieving the ontology model (810). The final result will be returned to the requester through the Workflow Manager (812).

FIG. 9.0 is a flowchart illustrating the steps for creating rules for inference engine in the Rule-Based Inference Engine. Ontology will be pre-configured with customized rules based on business cases. Request and trigger is received from Workflow Manager (902). Rule is identified (904) and said rule is invoked (906). Thereafter, ontology is accessed and updated (908). The result is obtained by retrieving the ontology model (910). The final result will be returned to requester through the Workflow Manager (912).

FIG. 10.0 is a flowchart illustrating the steps for integrating data through ontology in the Data Integrator. Ontology will be pre-configured with pre-defined data sources (either from internal or external).

Thereafter, data will be mapped in Ontology Model for decision making and the final result will be returned to the requester. Request and trigger is received from Workflow Manager (1002). Data integration type is resolved (1004) by determining selection of data integration type (1006). If data integration is of output type, ontology model is accessed; ontology is written to external data source; and returned to Workflow Manager. If data integration is of input type, external data source is mapped and created to ontology. The final result will be returned to requester through the Workflow Manager.

The present invention utilizes ontology for data representation, processes and the relationship of the subsystems based on an open standard, thus, its solutions are reusable by other open standard, ontology-based system. Further, the system defines a generic method for external application to make request to and receive answer from the system which provides flexibility to send input data in variety of lengths and formats and to pick up by the system for decision making. The answer return by the system is flexible as it could be configured by the user via ontology to return the answer in Boolean type, in quantitative figure, in qualitative rate, in string format, etc.

Unless the context requires otherwise or specifically stated to the contrary, integers, steps or elements of the invention recited herein as singular integers, steps or elements clearly encompass both singular and plural forms of the recited integers, steps or elements.

Throughout this specification, unless the context requires otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated step or element or integer or group of steps or elements or integers, but not the exclusion of any other step or element or integer or group of steps, elements or integers. Thus, in the context of this specification, the term "comprising" is used in an inclusive sense and thus should be understood as meaning "including principally, but not necessarily solely".

It will be appreciated that the foregoing description has been given by way of illustrative example of the invention and that all such modifications and variations thereto as would be apparent to persons of skill in the art are deemed to fall within the broad scope and ambit of the invention as herein set forth.

Claims

1. A system (100) for ontology-based decision support which utilizes open standards and information reuse comprising:

at least one Workflow Manager (104) for managing workflow of predefined solution;

at least one Data Integrator (1 10) for interacting with variety of information from a plurality of sources;

at least one Rule-based Inference Engine (106) for supporting automatic decision making based on pre-defined rules that interoperate with a range of rule-based reasoners; and at least one Algorithm Library (108) for processing algorithm to enable deployment of algorithms based on methodology defined in said Library without any modification of the ontology; and said algorithms are shared between heterogeneous implementers wherein each implementer is identified through a unique identification (ID) which is defined in the system.

2. A system (100) according to Claim 1 , wherein the rule-based reasoners are in plug and play manner.

3. A system (100) according to Claim 1 , wherein the at least one Algorithm Library (108) is an Algorithm Library.

4. A system (100) according to Claim 1 , wherein the at least one Algorithm Library (108) further comprising a lookup directory, said lookup directory is used to load and execute algorithms in said Library.

5. A system (100) according to Claim 1 , wherein said system (100) decouples the at least one Workflow Manager (104), the at least one Data Integrator (1 10), the at least one Rule-based Inference Engine (106) and the at least one Algorithm Library (108) in said system together with information required for deployment of decision support.

6. A method (400) for ontology-based decision support which utilizes open standards and information reuse comprising steps of:

retrieving information from heterogeneous sources (402);

predefining a set of semantic rules (404);

mapping data values among semantic rules (406);

configuring workflow and processes in ontology (408); and

integrating data through ontology (410)

characterized in that

integrating data through ontology further comprises steps of:

receiving request and trigger from Workflow Manager (1002);

resolving data integration type (1004);

determining selection of data integration type (1006) ;

if data integration is of output type:

accessing ontology model (1010); writing ontology to external data source (1012); and

returning to Workflow Manager (1014)

if data integration is of input type:

mapping and creating external data source to ontology (1008).

7. A method (400) according to Claim 6, wherein configuring workflow and processes in ontology (408) further comprises steps of:

accepting request (702) and managing workflow of a predefined solution (704); generating ontology model (706);

identifying process (708) and executing process (710);

updating ontology model (712); and

determining if process ended (714)

if process has not end:

reiterating step 708;

if process ended:

retrieving ontology model (716); and

constructing (718) and returning response (720).

8. A method (400) according to Claim 6, wherein configuring workflow and processes in ontology (408) further comprises attaching algorithms to the processes and creating rules for inference engine

9. A method (400) according to Claim 8, wherein attaching algorithms to processes further comprises steps of:

receiving request and trigger from Workflow Manager (802);

identifying algorithm (804) and invoking said algorithm (806);

accessing and updating ontology (808);

retrieving ontology model (810); and

returning to Workflow Manager (812).

10. A method (400) according to Claim 8, wherein creating rules for inference engine further comprises steps of:

receiving request and trigger from Workflow Manager (902);

identifying rule (904) and invoking said rule (906);

accessing and updating ontology (908);

retrieving ontology model (910); and

returning to Workflow Manager (912).