3.2 Adaptation Characteristics and Taxonomy
Christian et al. [21] presented a taxonomy of the different properties of self-adaptive
software. We will analyze this work and do a projection on our system requirements
and use the results to build our system.
Time: Handte et al. [22] provided two perspectives of temporal aspects: (i) Reactive is
when we have to adapt whenever a change in the context does happen. (ii) Proactive is
when the monitored data is used to forecast system behavior or environmental state
[21]. In our case, the adaptation will be reactive depending on the changes that happen
in the user contextual data.
Reason: The adaptation could be triggered for three reasons: (i) change of the context,
(ii) change in the technical resources, and (iii) change in the users. In our case, the
adaptation is triggered due to contextual changes, which provide a potential solution for
the multiscale nature of COPD.
Level: In our system, the change needs to be done on the application layer, where we
need to update the acceptable range for the different datasets or we need to activate new
components or call new functions.
Technique: McKinley [23] provided two techniques for adaptive software: parameter
adaptation and compositional adaptation. Parameter adaptation achieves a modified
system behavior by adjusting system parameters. Whereas compositional adaptation
enables the exchange of algorithms or system components dynamically at runtime. We
will use the first approach because it is suitable for a rule-based system.
Adaptation Control: Two approaches for implementing the adaptation logic can be
found in the literature. The internal approach, which twists the adaptation logic with the
system resources. The external approach splits the system into adaptation logic and
managed resources, The IBM Autonomic Computing Initiative provided MAPE Model
[24], which is an external, feedback control approach. Another aspect of the adaptation
logic is the degree of decentralization. We will follow a decentralized approach by
implementing independent units that control different aspects of adaptation.
4 Self-adaptation Healthcare System for COPD
4.1 Proposed System
Ajami and Mcheick [20] have proposed an ontology-based approach to keep track of
the physical status of patients, suggest recommendations and deliver interventions
promptly, by developing a decision support system based on an ontological formal
description that uses SWRL rules. The main goal of this paper is to provide an
adaptation architecture design for the application layer, which will address the connection between three different entities:
310
H. Mcheick et al.
Christian et al. [21] presented a taxonomy of the different properties of self-adaptive
software. We will analyze this work and do a projection on our system requirements
and use the results to build our system.
Time: Handte et al. [22] provided two perspectives of temporal aspects: (i) Reactive is
when we have to adapt whenever a change in the context does happen. (ii) Proactive is
when the monitored data is used to forecast system behavior or environmental state
[21]. In our case, the adaptation will be reactive depending on the changes that happen
in the user contextual data.
Reason: The adaptation could be triggered for three reasons: (i) change of the context,
(ii) change in the technical resources, and (iii) change in the users. In our case, the
adaptation is triggered due to contextual changes, which provide a potential solution for
the multiscale nature of COPD.
Level: In our system, the change needs to be done on the application layer, where we
need to update the acceptable range for the different datasets or we need to activate new
components or call new functions.
Technique: McKinley [23] provided two techniques for adaptive software: parameter
adaptation and compositional adaptation. Parameter adaptation achieves a modified
system behavior by adjusting system parameters. Whereas compositional adaptation
enables the exchange of algorithms or system components dynamically at runtime. We
will use the first approach because it is suitable for a rule-based system.
Adaptation Control: Two approaches for implementing the adaptation logic can be
found in the literature. The internal approach, which twists the adaptation logic with the
system resources. The external approach splits the system into adaptation logic and
managed resources, The IBM Autonomic Computing Initiative provided MAPE Model
[24], which is an external, feedback control approach. Another aspect of the adaptation
logic is the degree of decentralization. We will follow a decentralized approach by
implementing independent units that control different aspects of adaptation.
4 Self-adaptation Healthcare System for COPD
4.1 Proposed System
Ajami and Mcheick [20] have proposed an ontology-based approach to keep track of
the physical status of patients, suggest recommendations and deliver interventions
promptly, by developing a decision support system based on an ontological formal
description that uses SWRL rules. The main goal of this paper is to provide an
adaptation architecture design for the application layer, which will address the connection between three different entities:
310
H. Mcheick et al.