Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Open accessThe DOI matches its Crossref record; 48 of the 49 references it cites were found in a registry or an index; 8 of the 23 sources its claims cite were checked for support.
Abstract
Introduces DeepConvLSTM, a deep learning architecture combining convolutional and LSTM recurrent units for multimodal wearable activity recognition, outperforming non-recurrent deep networks and handcrafted feature baselines by 4% to 9%.
Overview
CoChat Summary
TL;DR
The paper proposes DeepConvLSTM, a deep framework combining four convolutional layers with two LSTM recurrent layers for wearable activity recognition from raw multimodal sensor data. It reports improved F1 scores over a baseline CNN and prior results on the OPPORTUNITY and Skoda datasets, including 0.915 F1 for OPPORTUNITY gestures with the Null class and 0.958 F1 on Skoda.
Key findings
- On OPPORTUNITY, DeepConvLSTM achieved F1 scores of 0.930 for modes of locomotion without Null, 0.895 for modes with Null, 0.866 for gestures without Null, and 0.915 for gestures with Null; the baseline CNN scored 0.912, 0.878, 0.783, and 0.883, respectively. The paper reports a 6% average improvement over the best OPPORTUNITY challenge submissions and more than 9% improvement on gesture recognition without Null.
- On Skoda, DeepConvLSTM reached 0.958 F1, compared with 0.884 for the baseline CNN, 0.893 for CNN [43], and 0.861 for CNN [23], improving the best reported result by 6%.
- Multimodal fusion improved gesture recognition on OPPORTUNITY without Null: 0.689 F1 with accelerometers only, 0.611 with gyroscopes only, 0.745 with accelerometers + gyroscopes, 0.839 with accelerometers + gyroscopes + magnetic sensors, and 0.864 with the full OPPORTUNITY sensor set.
- DeepConvLSTM had 999,122 parameters versus 7,445,458 for the baseline CNN, a 600% increase for the baseline; training took 340.3 min versus 282.2 min, and classification took 6.68 s versus 5.43 s, with DeepConvLSTM classifying almost 15 min of data per second and achieving 900× real-time on the GPU used.
Methods
- DeepConvLSTM uses four convolutional layers with 64 feature maps each, followed by two dense LSTM layers with 128 cells and a softmax output; the baseline CNN shares the convolutional layers but uses two non-recurrent dense layers with 128 units. Models were trained supervised with cross-entropy loss, mini-batch gradient descent with RMSProp, learning rate 10e-3, decay factor 0.9, mini-batches of 100 segments, and dropout p=0.5 on dense-layer inputs.
- Inputs were sliding windows of 500 ms with 250 ms step size, using raw sensor data with minimal preprocessing: linear interpolation for missing values and per-channel normalization to [0,1]. Evaluation used weighted F1 to address class imbalance.
- Experiments used the OPPORTUNITY dataset (on-body sensors, 113 channels, 30 Hz; Task A: 5-class modes of locomotion/postures; Task B: 18-class sporadic gestures) and the Skoda dataset (one subject, 10 right-arm 3D accelerometers, decimated to 30 Hz, 10 manipulative gestures, about 3 h and about 70 repetitions per gesture). Analyses included comparison to baseline CNN and published results, confusion matrices, sensor-modality fusion, and hyperparameter evaluations of sequence length and number of convolutional layers.
Limitations
- The paper states that DeepConvLSTM has not yet been implemented on a wearable device; the GPU used clearly outperforms current high-end wearable computational power, and training is best envisioned server-side. It also states that pooling was omitted because the sliding-window setup limits downsampling, though pooling could be useful without that requirement, and that training time increases with the number of layers, creating a performance–computational budget trade-off.
- Evident from the text: Skoda evaluation uses one subject and only right-arm sensors, and OPPORTUNITY testing is limited to ADL4 and ADL5 for Subjects 2 and 3, which constrains generalizability. The text does not report cross-validation or statistical significance testing.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures



![Figure 4 Placement of on-body sensors used in the OPPORTUNITYdataset (left: inertial measurements units; right: 3-axis accelerometers) [7].](https://figures.cochat.ai/paper-figures/27c0e4b39d104fe2b5315587d736fd2d/deep-convolutional-and-lstm-recurrent-neural-networks-for-multimodal-wearable-figure-4.jpg)




Verification
Claims the paper makes
5 Valid · 5 Not valid · 13 Not checked · 99 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“The choice of the number of recurrent layers is made following the results presented in [33], where the authors showed that a depth of at least two recurrent layers is beneficial when processing sequential data.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.1. DeepConvLSTM
What the sources say
- Supportedcites [33] Karpathy A. Johnson J. Li F.F. Visualizing and understanding recurrent networks Cornell Univ. Lab. 2015 arXiv:1506.02078 source
Section 4.1 compares recurrent networks with one, two, and three layers on two sequence datasets and states that a depth of at least two is consistently beneficial. This directly supports the attributed finding, within the paper’s tested settings. (LLM, conf=0.94)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“This seems to indicate that although deep learning techniques are often employed with large amounts of data, (e.g., millions of frames in computer vision [22]), they may actually be applicable to problem domains where acquiring annotated data is very costly, such as in supervised activity recognition.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 5. Results and Discussion > 5.4. Discussion
What the sources say
- Supportedcites [22] Pigou L. Oord A.V.D. Dieleman S. van Herreweghe M. Dambre J. Beyond Temporal Pooling: Recurrence and Temporal Convolutions for Gesture Recognition in Video arXiv Preprint 2015 arXiv:1506.01911 source
The citation is attached to the example of large-scale computer-vision data, not to the citing authors’ conclusion about costly annotation. Section 4.1 (page unavailable) reports 1,720,800 labeled video frames in the gesture-recognition dataset, supporting the cited example of data on the order of millions of frames. (LLM, conf=0.86)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The baseline CNN also offers better results than the OPPORTUNITY submissions in the recognition of models of locomotion.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 5. Results and Discussion > 5.1. Performance Comparison
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“From the results, we can also see how the model is able to learn from signals obtained from accelerometers, gyroscopes and magnetometers, fusing them without requiring any specific preprocessing.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 6. Conclusions
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“(iv) it is applicable even if the gestures are longer than the observation window.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 5. Results and Discussion > 5.4. Discussion
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
84 more claims
Create a free account to see all 122 claims and to verify the 2 references still open.
- It may state two separate claims as one. 14
- Probably not a substantial claim. 23
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source doesn’t support the claim as written.
In the citing paper
“However, these complex network framework suffered from low ef ciency and can hardly meet real-time requirements in practice applications.”
From InnoHAR: A Deep Neural Network for Complex Human Activity Recognition (2019) · cites this paper as [19]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe cited source doesn’t support the claim as written.
In the citing paper
“However, due to the lack of filtration techniques applied to multimodal data, the system’s performance was low.”
From A Multimodal IoT-Based Locomotion Classification System Using Features Engineering and Recursive Neural Network (2023) · cites this paper as [15]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“and yield better performances than pure CNNs or LSTM networks [12]”
From Comparison of Feature Learning Methods for Human Activity Recognition Using Wearable Sensors (2018) · cites this paper as [12]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The proposed method outperforms other methods, such as those based on deep believe networks (DBNs) [10], SVM [10], and CNNs [13].”
From Deep Recurrent Neural Networks for Human Activity Recognition (2017) · cites this paper as [13]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The popular DeepConvLSTM model [7] can only afford four, moderately-sized convolution layers to learn effective data representations, and two LSTM layers for sequence modeling.”
From Complex Deep Neural Networks from Large Scale Virtual IMU Data for Effective Human Activity Recognition Using Wearables (2021) · cites this paper as [7]
Experimental — accuracy not yet measured
20 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture
“The length of the kernels is the same for every convolutional layer, being defined as Pl=5,∀l=2,⋯,5.”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 4. Experimental Setup > 4.2. Performance Measure
“Following the OPPORTUNITY challenge experimental setup, the length of the window is 500 ms, with a step size of 250 ms.”
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“| Total | 996,800 +(128×nc)+nc | 996,800 +(128×nc)+nc | 7,443,136+(128×nc)+nc | 7,443,136+(128×nc)+nc |”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.3. Model Implementation and Training
Reported 996800 · recomputed 1037888
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“| 6 | Wai,Waf,Wac,Wao: 7232×128 | 942,592 | W: 57,856×128 | 7,405,696 |”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.3. Model Implementation and Training
Reported 942592 · recomputed 3702784
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“| 7 | Wai,Waf,Wac,Wao: 128×128 | 33,280 | W: 128×128 | 16,512 |”
§ Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition > 3. Architecture > 3.3. Model Implementation and Training
Reported 33280 · recomputed 131072
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562 · cited 1×
“Other publications showed much deeper networks, such as “GoogLeNet”, which is a 27-layer deep neural network applied to image classification [45].”
BackgroundReducing the dimensionality of data with neural networks
Hinton, G. E., & Salakhutdinov, R. R. (2006) · Science · doi:10.1126/science.1127647 · Cited by 20,928
Speech recognition with deep recurrent neural networks
Graves, A., Mohamed, A.-R., & Hinton, G. (2013) · 2013 IEEE International Conference on Acoustics, Speech and Signal Processing · doi:10.1109/icassp.2013.6638947 · Cited by 8,893 · cited 1×
“Networks using LSTM cells have offered better performance than standard recurrent units in speech recognition, where they gave state-of-the-art results in phoneme recognition [20].”
BackgroundConvolutional networks for images, speech, and time series
LeCun, Y., & Bengio, Y. (1998) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.5555/303568.303704 · Cited by 4,231
The class imbalance problem: A systematic study1
Japkowicz, N., & Stephen, S. (2002) · Intelligent Data Analysis · doi:10.3233/ida-2002-6504 · Cited by 3,295
DeepPose: Human Pose Estimation via Deep Neural Networks
Toshev, A., & Szegedy, C. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.214 · Cited by 3,185 · cited 1×
“Deep CNNs have had a major impact on fields, like content recommendation [24], speech recognition [25] and in computer vision [26–28], where they have become the de facto standard approach.”
BackgroundOverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
Sermanet, P., Eigen, D., Zhang, X., Mathieu, M., Fergus, R., & LeCun, Y. (2013) · arXiv · doi:10.48550/arxiv.1312.6229 · Cited by 2,680 · cited 1×
“Deep CNNs have had a major impact on fields, like content recommendation [24], speech recognition [25] and in computer vision [26–28], where they have become the de facto standard approach.”
BackgroundLee, H., Grosse, R., Ranganath, R., & Ng, A. Y. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553453 · Cited by 2,412 · cited 1×
“Deep learning techniques have outperformed many conventional methods in computer vision [13] and audio classification [14].”
MethodA review of wearable sensors and systems with application in rehabilitation
Patel, S., Park, H., Bonato, P., Chan, L., & Rodgers, M. (2012) · Journal of NeuroEngineering and Rehabilitation · doi:10.1186/1743-0003-9-21 · Cited by 2,396 · cited 1×
“, from simple hand gestures to complex activities, such as “cooking a meal”) and the context in which they occur from sensor data is at the core of smart assistive technologies, such as in smart homes [1], in rehabilitation [2], in health support [3,4], in skill assessment [5] or in industrial settings [6].”
Background
Research works printed but not linked: 1
- Gers F.A. Schraudolph N.N. Schmidhuber J. Learning precise timing with LSTM recurrent networks J. Mach. Learn. Res. 2003 3 115 143
Cited by
Items 1,401–1,500 of 2,807
An Optimal Feature Selection Method for Human Activity Recognition Using Multimodal Sensory Data
Haider, T., Khan, M. H., & Farid, M. S. (2024) · Information · doi:10.3390/info15100593 · Cited by 9
Rizk, H., Amano, T., Yamaguchi, H., & Youssef, M. A. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3176291 · Cited by 8
“A hybrid technique that harnesses LSTM in conjunction with a CNN for activity recognition is proposed in [41], denoted by DeepConvLSTM [41], to leverage their combined advantages.”
MethodDeep Transfer Learning Using Class Augmentation for Sensor-Based Human Activity Recognition
Kondo, K., & Hasegawa, T. (2022) · IEEE Sensors Letters · doi:10.1109/lsens.2022.3206472 · Cited by 9
“Several HAR methods are implemented by deep learning [2], [3], [4], which can achieve higher recognition performance than traditional machine learning methods that are based on hand-crafted features [4].”
MethodIn Shift and In Variance: Assessing the Robustness of HAR Deep Learning Models Against Variability
Khaked, A. A., Oishi, N., Roggen, D., & Lago, P. (2025) · Sensors · doi:10.3390/s25020430 · Cited by 14
“In contrast, hybrid models combine CNNs with sequential neural networks, such as RNNs [22], LSTMs [27], and GRUs [9].”
InfluentialSelf-citationLupión, M., Polo-Rodríguez, A., Medina-Quero, J., Sanjuan, J. F., & Ortigosa, P. M. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.117356 · Cited by 11
Akbari, A., & Jafari, R. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2020.3023109 · Cited by 8
“Another interesting conclusion from this figure is that hand-crafted features perform better than automated features of CNN which is in contrast to [21].”
MethodTFC: Time–frequency contrasting network for wearable-based human activity recognition
Huang, Z., Deng, J., Wang, S., Tang, C., & Xiao, S. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113373 · Cited by 9
Real Time Sign Language Recognition System for Hearing and Speech Impaired People
Petkar, T., Patil, T., Wadhankar, A., et al. (2022) · International Journal for Research in Applied Science and Engineering Technology · doi:10.22214/ijraset.2022.41765 · Cited by 8
Learning to Generate Rock Descriptions from Multivariate Well Logs with Hierarchical Attention
Tong, B., Klinkigt, M., Iwayama, M., Yanase, T., Kobayashi, Y., Sahu, A., & Vennelakanti, R. (2017) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3097983.3098132 · Cited by 7
Socialite: Social Activity Mining and Friend Auto-labeling
Wang, G., Zhang, L., Yang, Z., & Li, X. (2018) · IEEE International Performance, Computing, and Communications Conference · doi:10.1109/pccc.2018.8710834 · Cited by 10
“LSTMs contains two fully connected layers and two LSTM layers, and CRNN contains two convolutional layers, one fully connected layer and two LSTM layers.”
BackgroundMethodInfluentialEnd-to-end walking speed estimation method for smartphone PDR using DualCNN-LSTM.
Kawaguchi, N., Nozaki, J., Yoshida, T., Hiroi, K., Yonezawa, T., & Kaji, K. (2019) · Cited by 15
Ahmad, N., & Leung, H.-F. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3643511 · Cited by 13
“We follow a past HAR study [36] to choose kernel size as 5 for the temporal layer.”
BackgroundMethodZhang, K., Wang, Q., Meng, X., & Wang, J. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3380076 · Cited by 9
“…learning methods include random forest (RF) [35] and SVM [18], both of which are the most commonly used and effective machine learning methods in HAR Deep learning methods include DeepConvLSTM [32], AttnSense [27], hierarchical learning (HDL) [36], and deep long short-term memory (DLSTM) [37].”
MethodRobust Real-Time Group Activity Recognition of Robot Teams
Lu, L., Wang, H., Reily, B., & Zhang, H. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3060723 · Cited by 8
“Deep learning has also been applied recently, particularly for activity recognition from very noise input data, such as from wearable sensors [19], [20].”
BackgroundRen, Y., Liu, M., Yang, Y., Mao, L., & Chen, K. (2024) · Digital Health · doi:10.1177/20552076231223804 · Cited by 11
SONAR, a nursing activity dataset with inertial sensors
Konak, O., Döring, V., Fiedler, T., Liebe, L., Masopust, L., Postnov, K. V., Sauerwald, F., Treykorn, F., Wischmann, A., Kalabakov, S., et al. (2023) · Scientific Data · doi:10.1038/s41597-023-02620-2 · Cited by 9
Incept_LSTM : Accession for human activity concession in automatic surveillance
Girdhar, P., Johri, P., & Virmani, D. (2020) · Journal of Discrete Mathematical Sciences and Cryptography · doi:10.1080/09720529.2020.1804132 · Cited by 10
“Evaluation Metric Method 1 (LSTM Ensemble) [26] Method 2 (Two-stream convolutional networks [20] Proposed Method (Incept_ LSTM)”
BackgroundMethodLightHART: Lightweight Human Activity Recognition Transformer
Haque, S. T., Ni, J., Li, J., Yan, Y., & Ngu, A. H. H. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-78354-8_27 · Cited by 10
Jaén-Vargas, M., Leiva, K. R., Fernandes, F., Gonçalves, S., Silva, M., Lopes, D. S., & Olmedo, J. J. S. (2021) · Frontiers in artificial intelligence and applications · doi:10.3233/faia210196 · Cited by 11
CROMOSim: A Deep Learning-Based Cross-Modality Inertial Measurement Simulator
Hao, Y., Lou, X., Wang, B., & Zheng, R. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3230370 · Cited by 14
“We adopt the DeepConvLSTM network proposed in [44] as the task model, while the same simulator neural network trained on the TotalCapture dataset is used here to simulate sensor readings from videos.”
MethodActivities of Daily Living Classification using Recurrent Neural Networks
Jurca, R., Cioara, T., Anghel, I. M., et al. (2018) · RoEduNet Conference: Networking in Education and Research (RoEduNet) · doi:10.1109/roedunet.2018.8514124 · Cited by 8
“In [7] an implementation based on convolution neural networks and LSTM cells is presented.”
MethodA Wearable Fall Detection System Based on 1D CNN
Liu, P., Pan, J., Zhu, H., & Li, Y. (2021) · International Conference on Artificial Intelligence and Computer Engineering (ICAICE) · doi:10.1109/icaice54393.2021.00046 · Cited by 8
“Experimental results of reference [12] show that the sliding window of 2 seconds is better than the sliding window of 1, 3 seconds.”
BackgroundHuang, M., & Garcia, A. (2023) · Applied Sciences · doi:10.3390/app132011379 · Cited by 9
Human Activity Recognition Based on Transfer Learning
Pang, J. (2018) · Digital Commons - University of South Florida (University of South Florida) · Cited by 9
Multi-dimensional task recognition for human-robot teaming: literature review
Baskaran, P., & Adams, J. A. (2023) · Frontiers in Robotics and AI · doi:10.3389/frobt.2023.1123374 · Cited by 12
“Deep learning algorithms can recognize concurrent and composite fine-grained motor tasks directly from raw sensor data using complex network architectures, provided sufficient data is available (Ordóñez and Roggen, 2016; Zhao et al., 2018).”
MethodConvolutional Neural Network with an Elastic Matching Mechanism for Time Series Classification
Ouyang, K., Hou, Y., Zhou, S., & Zhang, Y. (2021) · Algorithms · doi:10.3390/a14070192 · Cited by 10
“It has a wide range of applications in human activity recognition [2], speech analysis [3], electrocardiogram (ECG) monitoring [4], and biological research [5].”
BackgroundLarge-Scale Continuous Mobility Monitoring of Parkinson’s Disease Patients Using Smartphones
Cheng, W., Lipsmeier, F., Creigh, A., Scotland, A., Kilchenmann, T., Jin, L., Schjodt‐Eriksen, J., Wolf, D., Zhang, Y., Garcia, I. F., et al. (2018) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-319-98551-0_2 · Cited by 8
A survey of cyber-physical system implementations of real-time personalized interventions
Steele, R. J., Hillsgrove, T., Khoshavi, N., & Jaimes, L. G. (2021) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-021-03263-0 · Cited by 8
Barrier Detection Using Sensor Data from Multiple Modes of Transportation with Data Augmentation
Kurauchi, Y., Abe, N., Konishi, H., & Seshimo, H. (2019) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac.2019.00101 · Cited by 8
“Sensing has been used to recognize user actions [3][4] and various other behaviors [5].”
BackgroundMulti-label Prediction in Time Series Data using Deep Neural NetworksPreprint
Zhang, W., Jha, D. K., Laftchiev, E., & Nikovski, D. N. (2020) · International Journal of Prognostics and Health Management · doi:10.48550/arxiv.2001.10098 · Cited by 7
A Deep Learning Method for Complex Human Activity Recognition Using Virtual Wearable SensorsPreprint
Xiao, F., Pei, L., Chu, L., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.01874 · Cited by 8
Li, Y., Dong, C., Jiang, C., et al. (2024) · Engineering Failure Analysis · doi:10.1016/j.engfailanal.2024.108994 · Cited by 8
Deep Recurrent Neural Network for Mobile Human Activity Recognition with High ThroughputPreprint
Inoue, M., Inoue, S., & Nishida, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.03607 · Cited by 8
“In concrete, the total size of the inner variables and architecture from the input layer to the output layer of the DRNN is 74166 pcs, and this is the size of less than 10% of the conventional CNN + LSTM model [24].”
BackgroundMethodInfluentialDetecting Face-Touch Hand Moves Using Smartwatch Inertial Sensors and Convolutional Neural Networks
Şehirli, E., & Alesmaeil, A. (2022) · International Journal of Intelligent Systems and Applications in Engineering · doi:10.18201/ijisae.2022.275 · Cited by 23
Kim, J. S., Nam, Y., Lee, J., Suh, Y.-J., & Hwang, I. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3610920 · Cited by 21
Deep stair walking detection using wearable inertial sensor via long short-term memory network
Nie, G. W., Ghazali, N. F., Shahar, N., & As’ari, M. A. (2020) · Bulletin of Electrical Engineering and Informatics · doi:10.11591/eei.v9i1.1685 · Cited by 8
“Recently, deep learning model has been formulated in HAR related studies [15-19] to overcome the limitation of common machine learning approaches.”
BackgroundClustering of user activities based on adaptive threshold spiking neural networks
Amin, H. H., Deabes, W. A., & Bouazza, K. (2017) · International Conference on Ubiquitous and Future Networks · doi:10.1109/icufn.2017.7993735 · Cited by 8
“Although several intelligent research have been suggested, they have limited performances due to the use of only the sensors reading values regardless the time of these readings [10]–[14].”
BackgroundTemporal Convolutional Network with Wavelet Transform for Fall Detection
Lu, X., Ling, Y., & Liu, S. (2022) · Journal of Sensors · doi:10.1155/2022/7267099 · Cited by 10
“[51] combined LSTM with batch normalization to achieve 92% prediction accuracy on raw accelerometer and gyroscope data, while Ordóñez and Roggen [58] proposed a hybrid CNN and LSTM model (DeepConvLSTM) for activity recognition based on data from multimodal wearable sensors.”
MethodGhosh, S., Kumar, S., Lepcha, S., et al. (2020) · Lecture notes in networks and systems · doi:10.1007/978-981-15-5309-7_27 · Cited by 8
Deep Neural Networks for Time Series Classification in Human Activity Recognition
Joshi, S., & Abdelfattah, E. (2021) · IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON) · doi:10.1109/iemcon53756.2021.9623228 · Cited by 9
Deep learning and low rank dictionary model for mHealth data classification
Said, A. B., Mohamed, A. M. S., Elfouly, T., Abualsaud, K., & Harras, K. A. (2018) · International Wireless Communications and Mobile Computing Conference (IWCMC) · doi:10.1109/iwcmc.2018.8450434 · Cited by 8
Investigating (re)current state-of-the-art in human activity recognition datasets
Bock, M., Hoelzemann, A., Moeller, M., & Laerhoven, K. V. (2022) · Frontiers in Computer Science · doi:10.3389/fcomp.2022.924954 · Cited by 10
Brooks, C., Atreya, M., & Szafir, D. J. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8594180 · Cited by 10
Deep Learning Shared Bandpass Filters for Resource-Constrained Human Activity Recognition
Stuart, M., & Manic, M. (2021) · IEEE Access · doi:10.1109/access.2021.3064031 · Cited by 8
ALGAN: Time Series Anomaly Detection with Adjusted-LSTM GAN
Bashar, M. A., & Nayak, R. (2023) · doi:10.21203/rs.3.rs-3609809/v1 · Cited by 22
Memiş, G., & Sert, M. (2019) · IEEE Sensors Journal · doi:10.1109/jsen.2019.2916393 · Cited by 9
“Ordóñez and Roggen [14] proposed”
Duc, P. T., & Toan, V. D. (2025) · Measurement Interdisciplinary Research and Perspectives · doi:10.1080/15366367.2025.2464970 · Cited by 8
“Due to its advantages in local dependence and invariance, CNN has been widely applied in activity recognition studies, particularly in human fall detection (Ordonex & Roggen, 2016; Rueda et al., 2018).”
Mehedi, S. T., Abdulrazak, L. F., Ahmed, K., Uddin, M. S., Bui, F. M., Chen, L., Moni, M. A., & Al-Zahrani, F. A. (2025) · Scientific Reports · doi:10.1038/s41598-025-89612-x · Cited by 9
Hierarchical Multi-Classification for Sensor-based Badminton Activity Recognition
Wang, Y., Ma, J., Li, X., & Zhong, A. (2020) · International Conference on Signal Processing · doi:10.1109/icsp48669.2020.9320935 · Cited by 9
“Human activity recognition (HAR) is a fundamental and practical task, which has attracted increasing research efforts in recent years due to its extremely broad applications in many areas, such as healthcare [1], gesture recognition [2]”
BackgroundHoomod, H. K., & Amory, Z. S. (2020) · International Conference on Communication and Electronics Systems (ICCES) · doi:10.1109/icces48766.2020.9137885 · Cited by 8
“In general deeper networks give better accuracy but in RNN do not go more depth[10] , In this type of neural network have challenging to learn sequences of data deal with real time [11].”
BackgroundOptimal sensor channel selection for resource-efficient deep activity recognition
Leite, C. F. S., & Xiao, Y. (2021) · ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN) · doi:10.1145/3412382.3458278 · Cited by 13
“For the PAMAP2, following [11, 19, 24, 25, 33], the window size contained approximately 5.”
HUMAN ACTIVITY CLASSIFICATION INCORPORATING EGOCENTRIC VIDEO AND INERTIAL MEASUREMENT UNIT DATA
Lu, Y., & Velipasalar, S. (2018) · IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2018.8646367 · Cited by 9
“For the first layer LSTM for the IMU data, D = 512, N = 36, and t ∈ [1, 16].”
BackgroundMethodInfluentialDynamic Driving Style Recognition for Human Machine Shared Control
Ye, P., Song, B., Yang, J., & Xiong, G. (2024) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2024.3490574 · Cited by 9
“The effectiveness of the classification was measured using the F1-measure [59]: where P is precision and P = TP TP + FP , R = TP TP + FN are precision and recall.”
Spectrum-Guided Adversarial Disparity Learning
Liu, Z., Yao, L., Bai, L., Wang, X., & Wang, C. (2020) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3394486.3403054 · Cited by 9
“• MC-CNN [28]: a state-of-the-art Convolutional Neural Network • Bi-LSTM-S [10]: a bi-directional Long Short-term Memory • ConvLSTM [18]: a hybrid model that combines both CNN and • En-LSTM [9]: an LSTM-based method that combines multiple • AttConvLSTM [17]: a hybrid model that applies attention…”
Wieland, F., & Nigg, C. (2023) · JMIR AI · doi:10.2196/42337 · Cited by 8
“Ordóñez and Roggen [15] presented a deep-CNN–based framework, which they tested against models such as decision tree, random forest, and support vector machines.”
MethodThe Human Continuity Activity Semisupervised Recognizing Model for Multiview IoT Network
Yuan, R., & Wang, J. P. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3234053 · Cited by 8
“Ordóñez and Roggen propose the Deep ConvLSTM [33] algo-rithm composed of convolutional and LSTM recurrent layers, which can automatically model the temporary dependency of sensor data.”
BackgroundMethodInfluentialImproving Wearable-Based Activity Recognition Using Image Representations
Guinea, A. S., Sarabchian, M., & Mühlhäuser, M. (2022) · Sensors · doi:10.3390/s22051840 · Cited by 12
“…best on the best results from a set of LSTM models [46], approaches that consider the machine learning concept of attention to improve performance [36,47], and an end-to-end approach that combines convolutional networks and LSTM models to account for the spatio-temporal patterns in the data [7].”
BackgroundMethodInfluentialAlbaba, M., Qassab, A., & Yılmaz, A. (2020) · International Journal of Applied Mathematics Electronics and Computers · doi:10.18100/ijamec.803105 · Cited by 8
“types like accelerometer, sound, luminance and other sensors [6-8].”
BackgroundFarid, F. A., Bari, A., Miah, A. S. M., Mansor, S. B., Uddin, J., & Kumaresan, S. P. (2025) · Journal of Imaging · doi:10.3390/jimaging11060182 · Cited by 16
Human Activity Recognition Based on a Modified Capsule Network
Zhu, S., Chen, W., Liu, F., Zhang, X., & Han, X. (2023) · Mobile Information Systems · doi:10.1155/2023/8273546 · Cited by 13
Prediction of Wax Deposits for Crude Pipelines Using Time-Dependent Data Mining
Yao, B., Chen, J., Li, C., et al. (2021) · SPE Journal · doi:10.2118/205374-pa · Cited by 8
Jiang, W., Wang, K., Lv, Y. A., Guo, J., Ni, Z., & Ni, Y. (2019) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2019.122884 · Cited by 9
GLULA: Linear attention-based model for efficient human activity recognition from wearable sensors
Bolatov, A., Yessenbayeva, A., & Yazıcı, A. (2024) · Wearable Technologies · doi:10.1017/wtc.2024.5 · Cited by 12
“DeepConvLSTM (Ordóñez and Roggen, 2016) (also known as DCL): A classical approach for HAR tasks that combines convolutional layers with recurrent units.”
BackgroundActivity-Aware Vital Sign Monitoring Based on a Multi-Agent Architecture
Ivaşcu, T., & Negru, V. (2021) · Sensors · doi:10.3390/s21124181 · Cited by 16
CORDIC Framework for Quaternion-based Joint Angle Computation to Classify Arm Movements
Biswas, D., Ye, Z., Mazomenos, E. B., et al. (2018) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas.2018.8350967 · Cited by 8
“Sensor-based activity recognition generally involves intensive computations in feature engineering/classification using learning algorithms [4] or Deep Learning [5].”
MethodSparse and dense matrix multiplication hardware for heterogeneous multi-precision neural networks
Nunez-Yanez, J. L., & Hosseinabady, M. (2021) · Array · doi:10.1016/j.array.2021.100101 · Cited by 14
Smartphone-Aided Human Activity Recognition Method using Residual Multi-Layer Perceptron
Shi, S., Wang, Y., Dong, H., Gui, G., & Ohtsuki, T. O. (2022) · IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS) · doi:10.1109/infocomwkshps54753.2022.9798274 · Cited by 9
Oh, B., Hwang, J., Seo, S., Chun, S., & Lee, K.-H. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.114912 · Cited by 9
The Effect of Axis-Wise Triaxial Acceleration Data Fusion in CNN-Based Human Activity Recognition
Han, X., Ye, J., Luo, J., & Zhou, H. (2020) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2018edp7409 · Cited by 9
“Ord´o˜nez and Roggen [25] combined convolutional and LSTM recurrent units to propose a generic framework for activity recognition.”
BackgroundALS-HAR: Harnessing Wearable Ambient Light Sensors to Enhance IMU-Based Human Activity Recognition
Ray, L. S. S., Geißler, D., Liu, M., Zhou, B., Suh, S., & Lukowicz, P. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-78110-0_9 · Cited by 12
UniMTS: Unified Pre-training for Motion Time Series
Zhang, X., Teng, D., Chowdhury, R., Li, S., Hong, D., Gupta, R., & Shang, J. (2024) · Advances in Neural Information Processing Systems 37 · doi:10.52202/079017-3414 · Cited by 8
Evaluation Model of Autonomous Vehicles’ Speed Suitability Based on Overtaking Frequency
Li, S., Huang, M., Guo, M., et al. (2021) · Sensors · doi:10.3390/s21020371 · Cited by 7
Skeleton Focused Human Activity Recognition in RGB VideoPreprint
Yu, B. X. B., Liu, Y., & Chan, K. C. C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.13979 · Cited by 8
“Multimodal HAR could be roughly categorized into two classes: vision-based multimodal [7-9] and vision-wearable-based multimodal [10-12].”
BackgroundDeep Learning for Sensor-Based Activity Recognition: Recent Trends
Ahad, M. A. R., Antar, A. D., & Ahmed, M. (2020) · Intelligent systems reference library · doi:10.1007/978-3-030-51379-5_9 · Cited by 7
Takata, T., Sasaki, H., Yamano, H., Honma, M., & Shikano, M. (2021) · doi:10.1101/2021.02.25.21252496 · Cited by 11
Stick-Slip Classification Based on Machine Learning Techniques for Building Damage Assessment
Na, Y., El‐Tawil, S., Ibrahim, A., & Eltawil, A. M. (2021) · Journal of Earthquake Engineering · doi:10.1080/13632469.2021.1891156 · Cited by 8
“In addition, a generic framework for human activity recognition based on convolutional and LSTM recurrent units was proposed (Ordóñez and Roggen 2016).”
Emotions During Covid-19: LSTM Models for Emotion Detection in Tweets
Javed, N., & Muralidhara, B. L. (2022) · Lecture notes in networks and systems · doi:10.1007/978-981-16-6407-6_13 · Cited by 8
Waveglove: Transformer-Based Hand Gesture Recognition Using Multiple Inertial Sensors
Králik, M., & Šuppa, M. (2021) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco54536.2021.9616000 · Cited by 12
Spatiotemporal analysis using deep learning and fuzzy inference for evaluating broiler activities
Chang, C., Xie, B.-X., & Xu, R.-Y. (2024) · Smart Agricultural Technology · doi:10.1016/j.atech.2024.100534 · Cited by 7
Vision and Acceleration Modalities: Partners for Recognizing Complex Activities
Diete, A., Sztyler, T., & Stuckenschmidt, H. (2019) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomw.2019.8730690 · Cited by 12
“Finally, for classification, classifiers that are commonly used are Decision Trees [18], Hidden Markov Models [21], and kNN [22] and recently Neural Networks [23].”
MethodEnd-to-End Learning for Visual Navigation of Forest Environments
Niu, C., Zauner, K., & Tarapore, D. (2023) · Forests · doi:10.3390/f14020268 · Cited by 7
“In summary, hand-crafted features are impaired by engineering bias and often lead to poor discriminative power [27].”
Jeon, S., Lee, Y., & Son, S. H. (2023) · IEEE Access · doi:10.1109/access.2023.3301872 · Cited by 8
Indumathi, V. M., & Prabakeran, S. (2021) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-981-16-0965-7_70 · Cited by 8
A Review of AI in Human-Machine Cooperation: Machine Perspective
Sourav, M. S. G., Javaid, A. Y., & Cheng, L. (2025) · ACM Transactions on Autonomous and Adaptive Systems · doi:10.1145/3774318 · Cited by 13
Sync or Sink? The Robustness of Sensor Fusion Against Temporal Misalignment
Kuhse, D., Hölscher, N., Günzel, M., Teper, H., von der Brüggen, G., Chen, J.-J., & Lin, C. (2024) · IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS) · doi:10.1109/rtas61025.2024.00018 · Cited by 19
“One classical, non-neural network specific approach is to preprocess the data before feeding it into the fusion function to reduce noise and other perturbations, such as interpolating to fill in missing data [40].”
BackgroundMulti-Headed Deep Learning Models to Detect Abnormality of Alzheimer’s Patients
Ammal, S. M., & Manoharan, P. S. (2022) · Computer Systems Science and Engineering · doi:10.32604/csse.2023.025230 · Cited by 6
Assessment of Hull and Propeller Performance Degradation Based on TSO-GA-LSTM
Huang, G., Liu, Y., Xin, J., et al. (2024) · Journal of Marine Science and Engineering · doi:10.3390/jmse12081263 · Cited by 7
Multi-occupant tracking with radar and wearable devices for enhanced accuracy in indoor environments
Polo-Rodríguez, A., Anguita-Molina, M. Á., Rojas, I., & Medina-Quero, J. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110872 · Cited by 12
Wang, X., He, M., Liu, Y., Wang, H., & Zhong, Y. (2022) · Electronics · doi:10.3390/electronics12010050 · Cited by 12
Use of Wearable Technologies for Analysis of Activity recognition for sports
Sharma, A., Al-Dala’in, T., ALSadoon, G. M. W., & Alwan, A. A. (2020) · International Conference on Innovative Technologies in Intelligent Systems and Industrial Applications (CITISIA) · doi:10.1109/citisia50690.2020.9371779 · Cited by 8
“b) Metabolic measurement Ord ́o ̃nez and Roggen [32] proposed wearable devices which have been used in the construction domain.”
MethodRecognition of Musical Instrument Using Deep Learning Techniques
Rajesh, S., & Nalini, N. J. (2021) · International Journal of Information Retrieval Research · doi:10.4018/ijirr.2021100103 · Cited by 7
Reddy, L., Anand, K., Kaushik, S., Rodrigo, C. J., McKay, J. L., Kesar, T. M., & Kwon, H. (2025) · PLOS Digital Health · doi:10.1371/journal.pdig.0001004 · Cited by 10
“While interpretable and efficient, these “shallow” methods lack the representational power of modern deep architectures such as DeepConvLSTMs [56], attention-based networks [57], Graph convolutional Networks, [58] or Transformer models [59].”
An, S., Gazi, A. H., & Inan, O. T. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3194677 · Cited by 8
“Deep convolutional-recurrent neural networks 100 (DeepConvLSTM) is a widely used architectural choice 101 in HAR using DNNs [18].”
MethodYadav, R. K., Arockiam, D., & Semwal, V. B. (2024) · Recent Advances in Computer Science and Communications · doi:10.2174/0126662558278156231231063935 · Cited by 7
CogAx: Early Assessment of Cognitive and Functional Impairment from Accelerometry
Ramamurthy, S. R., Chatterjee, S., Galik, E. M., Gangopadhyay, A., Roy, N., Mitra, B., & Chakraborty, S. (2022) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom53586.2022.9762401 · Cited by 11
Using Deep Learning and Smartphone for Automatic Detection of Fall and Daily Activities
Wu, X., Cheng, L., Chu, C., & Kim, J. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-34482-5_6 · Cited by 7
Multimodal fusion of IMUs and EPS body-worn sensors for scratch recognition
Jocys, Z., Pouryazdan, A., & Roggen, D. (2020) · International Conference on Pervasive Computing Technologies for Healthcare · doi:10.1145/3421937.3421987 · Cited by 7
“Moreover, to see what accuracy can be achieved for this dataset, investigating Deep Learning models such as DeepConvLSTM [12], could produce substantially higher accuracy results.”
BackgroundSelf-citationLSTM-CNN Architecture for Construction Activity Recognition Using Optimal Positioning of Wearables
Piyush, Rajak, S. K., & Vimal, K. E. K. (2024) · Journal of Construction Engineering and Management · doi:10.1061/jcemd4.coeng-14645 · Cited by 10
(2023) · doi:10.1101/2023.01.13.23284535 · Cited by 24
Smartphone Sensor Accelerometer Data for Human Activity Recognition Using Spiking Neural Network
Universiti Tun Hussein Onn Malaysia, M., Suriani, N. S., & Rashid, F. ‘. N. (2021) · International Journal of Machine Learning and Computing · doi:10.18178/ijmlc.2021.11.4.1051 · Cited by 9
Metadata
- Authors
- Ordóñez, F., & Roggen, D.
- Affiliations
- University of Sussex (GB)
- Corresponding author
- Ordóñez, Francisco
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2016-01-18
- Volume
- 16
- Issue
- 1
- Pages
- 115
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC4732148
- OpenAlex
- W2270470215
- Cited by
- 2,761
- References
- 49
- Citations by year
- 2026: 284 · 2025: 256 · 2024: 284 · 2023: 270 · 2022: 319 · 2021: 379 · 2020: 347 · 2019: 288 · 2018: 219 · 2017: 93 · 2016: 21
- FWCI
- 140.52
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Sussex Research Online (University of Sussex) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Figshare (submittedVersion, cc-by)
- Funders
- Google Faculty Research Award (Is deep learning useful for wearable activity recognition?)
- EPSRC (EP/N007816/1)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Engineering › Biomedical Engineering › Gait Recognition and Analysis
- Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
- Keywords
- human activity recognition, wearable sensors, multimodal sensor fusion, convolutional neural networks, LSTM neural network, temporal dynamics, automatic feature extraction, hyperparameter optimization
- MeSH terms
- Machine Learning, Clothing, Humans, Human Activities, Signal Processing, Computer-Assisted, Databases, Factual, Neural Networks, Computer, Monitoring, Ambulatory
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.