Selection of an Efficient Classification Algorithm for Ambient Assisted Living: Supportive Care for Elderly People

Area of Science:

• Gerontology
• Health Informatics
• Computer Science

Background:

• Ambient Assisted Living (AAL) systems integrate technology for elderly health and independence.
• Selecting effective classifiers is crucial for successful AAL technology adoption and user acceptance.
• Existing classification algorithms present challenges for decision-makers in technology acceptance modeling.

Purpose of the Study:

• To identify the optimal classification algorithm for supporting Ambient Assisted Living (AAL) implementation.
• To apply a multicriteria decision-making (MCDM) approach for evaluating technology acceptance classifiers.
• To guide decision-makers in selecting the most suitable classifier for AAL systems.

Main Methods:

• Utilized a multicriteria decision-making (MCDM) method.
• Employed the fuzzy method for order of preference by similarity to ideal solution (TOPSIS) for classifier evaluation.
• Compared multiple classification algorithms for their suitability in AAL.

Main Results:

• The K-Nearest Neighbors (KNN) classification algorithm was identified as the preferred technique.
• KNN demonstrated superior performance among the evaluated classification algorithms for AAL.
• The fuzzy TOPSIS MCDM method effectively ranked classifiers for AAL adoption.

Conclusions:

• KNN is the recommended classification algorithm for advancing Ambient Assisted Living (AAL) research and implementation.
• The MCDM approach using fuzzy TOPSIS provides a robust framework for classifier selection in technology acceptance.
• This study aids in optimizing AAL system development by identifying the best-performing classifier.