Decision support systems for home monitoring applications: Classification of activities of daily living and epileptic seizures

Luca, Stijn; Vuegen, Lode; hamme, Hugo Van; Karsmakers, Peter; Vanrumste, Bart

doi:10.1049/pbhe002e_ch13

Cited by 2 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A variety of classification algorithms have been applied to the problem of HAR, such as decision trees [10], [11], Naïve Bayes [10], Bayesian Networks [4], KNN [10], [11], convolutional neural networks [12], support vector machines (SVMs) [13], [3], and hidden Markov models (HMMs) [14]. Furthermore, deep learning approaches have recently gained much research attention [15], [16] and have been applied for HAR using low-power wearable devices [17], [18].…”

Section: A State-of-the-artmentioning

confidence: 99%

See 1 more Smart Citation

A Localized Learning Approach Applied to Human Activity Recognition

et al. 2021

Self Cite

View full text Add to dashboard Cite

The recognition of human physical activities and postures based on sensor data has received much research attention in several human health and biomedical engineering applications. In this study, the challenges of class-imbalance and ambiguity (or confusion) are discussed that frequently arise in data from human activity recognition (HAR) systems. In order to reduce the influence of imbalance and ambiguity in HAR problems, a novel hybrid localised learning approach of K-nearest neighbours leastsquares support vector machine (KNN-LS-SVM) is proposed. The classifier is applied to different synthetic and real-world datasets where imbalance and ambiguity are present. In this study, it is novel to apply a hybrid localised learning algorithm to the HAR problem. When compared to different global and local approaches, higher classification performances could be obtained by using the proposed localised learning approach. Furthermore, the computational effort could be reduced in an online learning mode.

show abstract

Section: A State-of-the-artmentioning

confidence: 99%

“…The importance of HAR systems is illustrated by the various amount of applications where they are used, e.g. medical monitoring [2], healthcare [3], military training, and sports [4]. Moreover, the use of HAR algorithms is enhanced by rapid advancements in sensor technology that enable to monitor people in their daily environments, e.g.…”

Section: Introductionmentioning

confidence: 99%

A Localized Learning Approach Applied to Human Activity Recognition

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…An imbalanced dataset is one in which the number of data samples is unbalanced between the different classes, where the number of samples occupies more in the majority class and less in the minority class. Imbalanced datasets are prevalent in real-world data, e.g., in credit card fraud detection [1], software defect detection [2], and cancer diagnosis [3]. The imbalance problem is common in the food domain since large amounts of qualified food data are often mixed with small amounts of unqualified food data, leading to imbalanced food datasets.…”

Section: Introductionmentioning

confidence: 99%