VARIABILITY MODELING METHOD FOR ROLE-BASED ARCHITECTURE IN ONTOLOGY-BASED SELF-ADAPTIVE SYSTEM

• US 20160371602A1
• Filed: 11/10/2015
• Published: 12/22/2016
• Est. Priority Date: 06/18/2015
• Status: Active Grant

• Current Assignee: Ajou University Industry
• Original Assignee: Ajou University Industry
• Inventors: LEE, Seokwon, PARK, Sinyoung
• Granted Patent: US 10,204,304 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition

Abstract

In a variability modeling method implemented in a computer system to implement a self-adaptive system, the variability modeling method includes building ontology in which a target system to be modeled is defined through requirement analysis of the target system, deciding whether adaptation is required by continuously monitoring the target system and a change in environment, and when it is decided that the adaptation is required, performing reconfiguration of the target system by determining the variable point suitable for a change in context.

Claims

1. A variability modeling method implemented in a computer system to implement a self-adaptive system, the variability modeling method comprising:

building ontology in which a target system to be modeled is defined through requirement analysis of the target system, wherein, in the ontology, a behavior feature required to achieve a goal of the target system and a component feature to perform the behavior feature are related to roles and players of a role-based architecture, and a variable point is defined in their relation;

deciding whether adaptation is required by continuously monitoring the target system and a change in environment; and

when it is decided that the adaptation is required, performing reconfiguration of the target system by determining the variable point suitable for a change in context.

2. The variability modeling method of claim 1, wherein, in the ontology, there are defined a goal model for defining a plurality of goals required to be achieved in the target system, a behavior feature model for defining at least one behavior required to achieve each goal, and a component feature model for defining at least one component to perform each behavior, and

wherein each feature in the behavior feature model and the component feature model is defined as a mandatory, optional, or alternative relation such that the variable point is expressed.

3. The variability modeling method of claim 2, wherein the goal model is defined based on a strategic rationale (SR) model of i*framework, including a goal of the target system, derived through the requirement analysis, an actor which is a subject to achieve the goal, and a task processed by the actor, and

wherein the behavior feature model is derived from the task derived from the goal model, and the component feature model is derived by the actor derived from the goal model.

4. The variability modeling method of claim 2, comprising when it is decided that the adaptation is required, determining, by a reasoning engine, the variable point through a query to the ontology.

5. The variability modeling method of claim 4, wherein the reasoning engine determines the variable point at which the reconfiguration of the target system is to be performed by changing a feature in the optional or alternative relation in each feature model.

6. The variability modeling method of claim 4, wherein, when mapping between the variable point is generated as the behavior feature and the role in the role-based architecture and mapping between the component feature and the player in the role-based architecture are performed through an n:

1, 1;

n, or n;

n relation, the reasoning engine determines the variable point at which the reconfiguration of the target system is to be performed by changing a role and a player in the mapping relation.

7. The variability modeling method of claim 4, comprising determining from which process the adaptation is to be performed, according to the determined variable point, among the process of changing the goal, the process of changing at least one feature among the behavior features and the component features, the process of changing the role, and the process of changing the player.