摘要：Recent Activity Daily Living (ADL) not only tackles simple activities, butalso caters to a wide range of complex activities. Although the same activity has beencarried out under the same environmental conditions, the acceleration signalobtained from each subject considerably differs. This happens due to the pattern ofaction generated for each subject is diverse based on several aspects such as subjectage, gender, emotion and personality. This project therefore compares the accuracyof various machine learning models for ADL classification. On top of that, thisresearch work also scrutinizes the effectiveness of various feature selection methodsto identify the most relevant attribute for ADL classification. As a result, RandomForest was able to achieve the highest accuracy of 83.3% on subject independentmatter in ADL classification. Meanwhile, CFS Subset Evaluator is considered to bea good feature selector as it successfully selected the 8 most relevant featurescompared with Correlation and Information Gain Evaluator.
关键词：Activity Daily Living (ADL); accelerometer; wearable sensor; machine
learning.
