Wearable sensor based multimodal human activity recognition exploiting the diversity of classifier ensemble

Guo, Haodong; Chen, Ling; Peng, Liangying; Chen, Gencai

doi:10.1145/2971648.2971708

Cited by 67 publications

(35 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Authors of [16] referred to the feature-level fusion as early fusion, which merges feature data extracted from each modality prior to the classification procedure. Using a single feature vector for the training model is straightforward, but feature compatibility issues regarding heterogeneous sampling frequencies and configuration parameters can deteriorate the classifier performance [35]. Contrarily, classifier-level fusion aims to combine the classification results given by the base-learners trained on different sensor modalities.…”

Section: Related Workmentioning

confidence: 99%

“…Contrarily, classifier-level fusion aims to combine the classification results given by the base-learners trained on different sensor modalities. As class-probability predictions are uniform in their expression, classifier-level ensemble learning techniques are widely adopted to integrate multiple machine learning models [34,35,36]. In this paper, we apply two-level stacking and voting ensembles, which fall into the classifier-level sensor fusion technique.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Sensor Data Acquisition and Multimodal Sensor Fusion for Human Activity Recognition Using Deep Learning

Chung

Lim

Noh

et al. 2019

Sensors

140

View full text Add to dashboard Cite

In this paper, we perform a systematic study about the on-body sensor positioning and data acquisition details for Human Activity Recognition (HAR) systems. We build a testbed that consists of eight body-worn Inertial Measurement Units (IMU) sensors and an Android mobile device for activity data collection. We develop a Long Short-Term Memory (LSTM) network framework to support training of a deep learning model on human activity data, which is acquired in both real-world and controlled environments. From the experiment results, we identify that activity data with sampling rate as low as 10 Hz from four sensors at both sides of wrists, right ankle, and waist is sufficient in recognizing Activities of Daily Living (ADLs) including eating and driving activity. We adopt a two-level ensemble model to combine class-probabilities of multiple sensor modalities, and demonstrate that a classifier-level sensor fusion technique can improve the classification performance. By analyzing the accuracy of each sensor on different types of activity, we elaborate custom weights for multimodal sensor fusion that reflect the characteristic of individual activities.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Sensor Data Acquisition and Multimodal Sensor Fusion for Human Activity Recognition Using Deep Learning

Chung

Lim

Noh

et al. 2019

Sensors

140

View full text Add to dashboard Cite

show abstract

“…• Lara et al [18] apply both statistical and structural detectors features to discriminate among activities. • Guo et al [12] exploit the diversity of base classifiers to construct a good ensemble for multimodal activity recognition, and the diversity measure is obtained from both labelled and unlabelled data. Note: If the compared method can not deal with unsupervised samples, it will be trained only by the supervised samples.…”

Section: Baselinesmentioning

confidence: 99%

Adversarial Variational Embedding for Robust Semi-supervised Learning

Zhang

Yao

Yuan

2019

Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

View full text Add to dashboard Cite

Semi-supervised learning is sought for leveraging the unlabelled data when labelled data is difficult or expensive to acquire. Deep generative models (e.g., Variational Autoencoder (VAE)) and semisupervised Generative Adversarial Networks (GANs) have recently shown promising performance in semi-supervised classification for the excellent discriminative representing ability. However, the latent code learned by the traditional VAE is not exclusive (repeatable) for a specific input sample, which prevents it from excellent classification performance. In particular, the learned latent representation depends on a non-exclusive component which is stochastically sampled from the prior distribution. Moreover, the semi-supervised GAN models generate data from pre-defined distribution (e.g., Gaussian noises) which is independent of the input data distribution and may obstruct the convergence and is difficult to control the distribution of the generated data. To address the aforementioned issues, we propose a novel Adversarial Variational Embedding (AVAE) framework for robust and effective semi-supervised learning to leverage both the advantage of GAN as a high quality generative model and VAE as a posterior distribution learner. The proposed approach first produces an exclusive latent code by the model which we call VAE++, and meanwhile, provides a meaningful prior distribution for the generator of GAN. The proposed approach is evaluated over four different real-world applications and we show that our method outperforms the state-of-the-art models, which confirms that the combination of VAE++ and GAN can provide significant improvements in semisupervised classification.

show abstract

“…Similar to [2], the experiments conducted on the two public datasets perform background activity recognition task [8]. The activities are categorized into 6 classes: lying, sitting/standing, walking, running, cycling and other activities.…”

Section: A Datasets and Experimental Settingsmentioning

confidence: 99%

“…To evaluate the performance of the proposed approach, our model, we conduct extensive experiments to compare its performance with the state-of-the-art methods on PAMAP2 and MHEALTH. We elaborately select other four state-of-theart and multimodal feature-based approaches (MARCEL [2], FEM [11], CEM [22] and MKL [23]) and five baseline methods (Support Vector Machine (SVM), Random Forest(RF), K-Nearest Neighbors(KNN), Decision Tree(DT) and Single Neural Networks) to show the competitive power of the proposed method. To ensure fair comparison, the best parameters test, our model, is used on both datasets; the best trade-off parameter (λ = 0.7) is deployed for MARCEL; time-domain features including mean, variance, standard deviation, median and frequency-domain features including entropy and spectral entropy are utilized for FEM; each modality feature group are defined an independent kernel for MKL; and for other baseline methods, all modality features are deployed.…”

Section: B Accuracy Comparison and Performance Analysismentioning

confidence: 99%

Interpretable Parallel Recurrent Neural Networks with Convolutional Attentions for Multi-Modality Activity Modeling

Chen

Yao

Wang

et al. 2018

2018 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Multimodal features play a key role in wearable sensor based human activity recognition (HAR). Selecting the most salient features adaptively is a promising way to maximize the effectiveness of multimodal sensor data. In this regard, we propose a "collect fully and select wisely" principle as well as an interpretable parallel recurrent model with convolutional attentions to improve the recognition performance. We first collect modality features and the relations between each pair of features to generate activity frames, and then introduce an attention mechanism to select the most prominent regions from activity frames precisely. The selected frames not only maximize the utilization of valid features but also reduce the number of features to be computed effectively. We further analyze the accuracy and interpretability of the proposed model based on extensive experiments. The results show that our model achieves competitive performance on two benchmarked datasets and works well in real life scenarios.

show abstract

Wearable sensor based multimodal human activity recognition exploiting the diversity of classifier ensemble

Cited by 67 publications

References 35 publications

Sensor Data Acquisition and Multimodal Sensor Fusion for Human Activity Recognition Using Deep Learning

Sensor Data Acquisition and Multimodal Sensor Fusion for Human Activity Recognition Using Deep Learning

Adversarial Variational Embedding for Robust Semi-supervised Learning

Interpretable Parallel Recurrent Neural Networks with Convolutional Attentions for Multi-Modality Activity Modeling

Contact Info

Product

Resources

About