2019/03/12 by Parviz Asghari, Asghari, Parviz, Elnaz Soelimani +3 · 1 citation
Computer Science · Engineering · Social Sciences · #Context-Aware Activity Recognition Systems #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #IoT and Edge/Fog Computing #IoT-based Smart Home Systems #Machine Learning (cs.LG) #Machine Learning (stat.ML)
paper · pdf · doi:10.48550/arxiv.1903.04820
openalex publication_date 2019/03/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In the last few years there has been a growing interest in Human Activity\nRecognition~(HAR) topic. Sensor-based HAR approaches, in particular, has been\ngaining more popularity owing to their privacy preserving nature. Furthermore,\ndue to the widespread accessibility of the internet, a broad range of\nstreaming-based applications such as online HAR, has emerged over the past\ndecades. However, proposing sufficiently robust online activity recognition\napproach in smart environment setting is still considered as a remarkable\nchallenge. This paper presents a novel online application of Hierarchical\nHidden Markov Model in order to detect the current activity on the live\nstreaming of sensor events. Our method consists of two phases. In the first\nphase, data stream is segmented based on the beginning and ending of the\nactivity patterns. Also, on-going activity is reported with every receiving\nobservation. This phase is implemented using Hierarchical Hidden Markov models.\nThe second phase is devoted to the correction of the provided label for the\nsegmented data stream based on statistical features. The proposed model can\nalso discover the activities that happen during another activity - so-called\ninterrupted activities. After detecting the activity pane, the predicted label\nwill be corrected utilizing statistical features such as time of day at which\nthe activity happened and the duration of the activity. We validated our\nproposed method by testing it against two different smart home datasets and\ndemonstrated its effectiveness, which is competing with the state-of-the-art\nmethods.\n