.png){kind=icon}
Return
X-OODM: Explainable Object-Oriented Design Methodology
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2024.3477553.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In software applications and decision-making systems, the explainability features can be instrumental for explicating internal working, accountability, understanding, fairness, and interpretation of decisions, processes, and data. Conventional design methodologies like Object-Oriented Design Methodology (OODM) are proposed for web-based application development. OODM enables the reuse of code, quantification, and security at the design level. However, OODM did not provide the feature of introducing explainability in web-based decision-making systems, thus OODM is required to be modified. The present paper presents X-OODM with an added model to introduce the explainability feature. Design quality metrics for X-OODM are also proposed. The proposed methodology is validated through a case study involving different scenarios. In the first scenario, trustworthiness, fairness, transferability, and simulatability are implemented, resulting in an explainability level of 24 units. In addition to these components, in the second scenario, reliability, understanding, informativeness, and decomposability are involved with the previous parameters, having an explainability level of 34 units. In the third scenario, in addition to defined parameters, privacy awareness, accessibility, and algorithmic transparency components are also implemented, leading to the highest level of explainability 46 units compared to previous scenarios. A higher explainability level indicates that all aspects of web-based applications introduce explainability. This research can be extended to implement X-OODM for a real multi-domain sentiment analysis application.
Keywords:
Object oriented modeling
Analytical models
Design methodology
Modeling
Decision making
Security
Sentiment analysis
Measurement
Numerical models
Privacy
Explainable AI
Explainable
measurable
web-based application
object-oriented design
sentiment analysis
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W
Organization
Cited Papers
ShareSaveShareSaveShareSaveExplainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI.png)
INFORMATION FUSION
OAAI
INFORMATION FUSIONIF15.5
OAAIShareSaveShareSave
View more