Human activity recognition (HAR) using smartphones is a hot research topic in computational intelligence. This paper reviews some research works in this decade and reports the advance from activity definition and recognition process, which includes data collection, data pre-processing, feature extraction, model training and learning, and model evaluation. Finally, this paper presents a summary and dis-cussion of some issues for further research on the smartphone activity recognition like activity description, data fusion of multiple sensors, data collection, feature and model selection, and training efficiency.
展开▼
