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 2,201–2,300 of 2,807
The Deep Learning-based Human Activity Recognition Using Smart Wearable Sensors : A Tutorial
Mekruksavanich, S., & Jitpattanakul, A. (2022) · Research Briefs on Information and Communication Technology Evolution · doi:10.56801/rebicte.v8i.133 · Cited by 1
“The resultant collection of forecasted labels is then evaluated using state-of-the-art HAR criteria (i.e., Accuracy, Recall, Precision, and F1-score).”
BackgroundMethodTernary Neural Networks for Gait Identification in Wearable Devices
Agnetti, G., Migliorati, A., Mari, D., Bianchi, T., Milani, S., & Magli, E. (2024) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs61860.2024.10810715 · Cited by 3
Qiang, Z., Zhang, Y., Haghighi, P. D., et al. (2021) · IEEE International Conferences on Internet of Things (iThings) and IEEE Green Computing & Communications (GreenCom) and IEEE Cyber, Physical & Social Computing (CPSCom) and IEEE Smart Data (SmartData) and IEEE Congress on Cybermatics (Cybermatics) · doi:10.1109/ithings-greencom-cpscom-smartdata-cybermatics53846.2021.00023 · Cited by 1
Liu, M., Zhou, B., Zhao, Z., Hong, H., Kim, H., Suh, S., Rey, V. F., & Lukowicz, P. (2023) · IEEE International Conference on Application-specific Systems, Architectures and Processors (ASAP) · doi:10.1109/asap57973.2023.00029 · Cited by 2
Activity Prediction Based on Deep Learning Techniques
Park, J., Song, C., Kim, M., & Kim, S. (2023) · Applied Sciences · doi:10.3390/app13095684 · Cited by 1
Systematic Evaluation of Deep Learning Models for Human Activity Recognition Using Accelerometer
Le, T.-H., Tran, Q.-H., & Le, T. (2020) · NAFOSTED Conference on Information and Computer Science (NICS) · doi:10.1109/nics51282.2020.9335853 · Cited by 2
“Precisely, three models that are Convolutional Neural Network (CNN) [1], DeepConvLSTM - a combination of CNN and Long Short Term Memory (LSTM) [2] and SensCapsNet - a Capsule Neural Network for wearable sensor-based HAR [3] were implemented and evaluated on three benchmark datasets that are…”
MethodInfluentialPacket Pruning: Finding Better Energy Spending Policies for Batteryless Human Activity Recognition
Cooper, G., & Marculescu, R. (2024) · Wearable and Implantable Body Sensor Networks · doi:10.1109/bsn63547.2024.10780463 · Cited by 1
(2020) · doi:10.1101/2020.07.08.20148015 · Cited by 1
“A CNN architecture with max pooling layers has shown varying results, from increased classification performance 33 to pooling layers interfering with the convolutional layer’s ability to learn to down sample the raw sensor data 34 .”
BackgroundMethodExploration of Sensor-Based Activity Recognition Based on Time Series Feature Extraction
Chen, W., Chen, T.-H., & Tsai, C.-H. (2021) · International Conference on Intelligent Systems and Image Processing (ICISIP) · doi:10.12792/icisip2021.023 · Cited by 1
Smartphone Gait Authentication Based on Activity Recognition Task in Unconstrained Environment
Cheng, D., Mao, R., Hao, Q., Wang, Y., & Li, J. (2024) · IEEE Symposium on Computers and Communications (ISCC) · doi:10.1109/iscc61673.2024.10733706 · Cited by 1
“To prove the beneficial impact of the optimized gait extraction model proposed in this paper on the entire authentication framework, it is compared with the deep learning model designed by [6], [7], [9], [10] and [17] in the field of activity recognition.”
BackgroundMethodIntelligent Calibration of Static FEA Computations Based on Terrestrial Laser Scanning Reference
Xu, W., Bao, X., Chen, G., & Neumann, I. (2020) · Sensors · doi:10.3390/s20226439 · Cited by 2
Contributions to Human Motion Modeling and Recognition using Non-intrusive Wearable Sensors
Martín, M. (2022) · UPM Digital Archive (Technical University of Madrid) · doi:10.20868/upm.thesis.70493 · Cited by 2
Mixing temporal experts for Human Activity Recognition
Roy, D., & Girdzijauskas, Š. (2022) · Swedish Artificial Intelligence Society Workshop (SAIS) · doi:10.1109/sais55783.2022.9833028 · Cited by 2
Arcolezi, H. H., Nunes, W. R. B. M., Cerna, S., de Araujo, R. A., Sanches, M. A. A., Teixeira, M. C. M., & de Carvalho, A. A. (2020) · Research on Biomedical Engineering · doi:10.1007/s42600-020-00089-1 · Cited by 1
“Similarly, Ordóñez and Roggen (2016) combined the LSTM with convolutional NNs for activity recognition from wearable sensors.”
BackgroundShojaedini, S. V., & Kharazi, M. H. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4655312 · Cited by 1
Yin, D., Xiao, K., Zou, C., Lu, Z., Jiao, C., Chen, M., He, R., Xu, Y., & Yang, P. (2026) · Measurement · doi:10.1016/j.measurement.2026.120572 · Cited by 1
Toward privacy-Aware human digital twins: A multi-Layer architecture
Lamazzi, L., Franco, F., & Bedogni, L. (2025) · Future Generation Computer Systems · doi:10.1016/j.future.2025.108263 · Cited by 4
Activity Recognition Analysis System and Application Based on Convolutional Neural Network
靳, 旭. (2020) · Computer Science and Application · doi:10.12677/csa.2020.109179 · Cited by 1
Data-Driven Cellular Network Selector for Vehicle Teleoperations
Gahtan, B., Cohen, R., Bronstein, A. M., et al. (2024) · International Conference on Network of the Future (NoF) · doi:10.1109/nof62948.2024.10741500 · Cited by 1
“The last two stages, (f) and (g), contain the proposed neural network architecture, which is inspired by [15].”
BackgroundA DeepConvLSTM Approach for Continuous Authentication Using Operational System Performance Counters
Andrade, C., Bragança, H. L. D. S., Fernandes, H., Feitosa, E. L., & Souto, E. (2025) · IET Biometrics · doi:10.1049/bme2/8262252 · Cited by 1
Recognizing Complex Activities by a Temporal Causal Network-Based Model
Liao, J., Hu, J., & Liu, L. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-67667-4_21 · Cited by 4
“DeepconvLSTM [22]: It infers complex activity directly from raw data.”
BackgroundWi-Fi Based Indoor Human Trajectory Tracking with Diffusion-Model
Zang, R. W., Yan, D., Wang, B., & Yan, Y. (2025) · IEEE International Conference on Parallel and Distributed Systems (ICPADS) · doi:10.1109/icpads67057.2025.11322924 · Cited by 1
An approach for Lane Keeping Assist at higher speeds using Deep Neural Network
Kumar, K. S., Adarsh, S., & Sangani, P. (2023) · IEEE Mysore Sub Section International Conference (MysuruCon) · doi:10.1109/mysurucon59703.2023.10397019 · Cited by 1
“This approach has been used for trajectory clustering [23], categorization of human activity [24], sentiment categorization [25], and other things.”
MethodDetecting Eating Events with Inertial Sensing in a Ring Wearable
Liang, D., Chiu, B., Arvonen, C., et al. (2026) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3789678 · Cited by 1
“In addition to CNNs, we also evaluated ADM performance using the DeepConvLSTM model [23], which incorporates additional LSTM layers to better capture the temporal dynamics of the input signals.”
A Machine Learning Framework for Activity Recognition With Robotic Ankle Foot Orthoses
Hopkins, S., Gaudette, C., Silveti, M. J., et al. (2026) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2026.3677751 · Cited by 1
“2) [30].”
Stankoski, S., Kiprijanovska, I., Gjoreski, M., Panchevski, F., Sazdov, B., Sofronievski, B., Cleal, A., Fatoorechi, M., Nduka, C., & Gjoreski, H. (2024) · doi:10.2196/preprints.59469 · Cited by 14
Smart Pallets: Towards Event Detection Using IMUs
Franke, S., Bommert, A., Brandt, M. J., Kuhlmann, J. L., Olivier, M.-C., Schorning, K., Reining, C., & Kirchheim, A. (2024) · IEEE International Conference on Emerging Technologies and Factory Automation (ETFA) · doi:10.1109/etfa61755.2024.10710674 · Cited by 2
Handle Dense Labeling in Human Activity Recognition Using Self Attention and BiLSTM
Thư, N. T., & Han, D. S. (2024) · International Conference on Consumer Electronics · doi:10.1109/icce59016.2024.10444332 · Cited by 2
“In case there is more than one type of activity in the window, the label will be decided by either the activity with the highest number of data samples [3] or the activity label of the last data sample in the window [4].”
BackgroundCumin, J., Ramparany, F., Lefebvre, G., et al. (2018) · Modélisation et utilisation du contexte · doi:10.21494/iste.op.2018.0207 · Cited by 1
Elderly Hajj pilgrims activity recognition based on candidate classification technique
Mohnish Karthikeyan, B., & Amutha, R. (2022) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.6932 · Cited by 3
In-Material Processing of High Bandwidth Sensor Measurements Using Modular Neural Networks
Hughes, D. (2018) · CU Scholar (University of Colorado Boulder) · Cited by 1
Continual Egocentric Activity Recognition via Foreseeable-Generalized Multi-Modal Representations
He, C., Cheng, S., Qiu, Z., Xu, L., Meng, F., Wu, Q., & Li, H. (2023) · doi:10.36227/techrxiv.24041583.v2 · Cited by 11
Predictive Wellness Analytics Through Hybrid Sensor-LLM Integration
Choudhari, R. D., & Khodwe, G. (2025) · International Conference on Big Data, Knowledge and Control Systems Engineering (BdKCSE) · doi:10.1109/bdkcse67969.2025.11300494 · Cited by 1
“Recent advances have incorporated deep learning approaches, with Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks showing superior performance for temporal activity patterns [12].”
Inferring travel activity pattern from smartphone sensing data using deep learning
Ghorpade, A. (2018) · DSpace@MIT (Massachusetts Institute of Technology) · Cited by 1
Investigation of Exponent-Free LSTM Cells for Virtual Sensing Applications
Jankauskas, M., Katkevičius, A., & Serackis, A. (2026) · Electronics · doi:10.3390/electronics15030576 · Cited by 1
Deep Auto-Set: A Deep Auto-Encoder-Set Network for Activity Recognition Using WearablesPreprint
Varamin, A. A., Abbasnejad, E., Shi, Q., Ranasinghe, D. C., & Rezatofighi, H. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.08127 · Cited by 3
“In this architecture, all convolution (and deconvolution) layers apply afilter ofwidth 5 (as in [16]) and”
BackgroundMethodInfluentialDetecting Door Operations Using Wearable Devices
Taniguchi, S., Ge, H., Hamada, T., Michikata, T., & Koshizuka, N. (2022) · IEEE Global Conference on Consumer Electronics (GCCE) · doi:10.1109/gcce56475.2022.10014368 · Cited by 1
“DeepConvLSTM [11] is a deep learning architecture used in HAR.”
Kumar, S., & Sharma, S. (2024) · Elsevier eBooks · doi:10.1016/b978-0-443-22158-3.00009-0 · Cited by 2
A Critical Review on Risk Assessment Methods of Musculoskeletal Disorder (MSD)
Siddhaiyan, V., Kumar, R. N., Ramya, P., Balasubramani, M., Sakthi, C., Sitheaswaran, C., Sandhiya, V. G., & Sakthivignesh, G. (2023) · Lecture notes in mechanical engineering · doi:10.1007/978-981-99-1665-8_20 · Cited by 2
Brain-Inspired Data Transmission in Dense Wireless Network
Kułacz, Ł., & Kliks, A. (2021) · Sensors · doi:10.3390/s21020576 · Cited by 1
An effective differentiable clustering approach for human activity recognition
Bouguelia, M.-R. (2026) · Advances in Data Analysis and Classification · doi:10.1007/s11634-026-00671-y · Cited by 1
Lang, L., Thuente, D. J., & Ma, X. (2023) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-231691 · Cited by 1
“T At present, the methods for measuring the physical strength consumption using motion sensors mainly focus on how to use IMUs to obtain the activity signals from different parts of the human body, and combine various data analysis models to improve measurement accuracy [12, 13].”
RetractedWang, J., Chen, Y., & Hu, C. (2023) · Springer eBooks · doi:10.1007/978-3-031-24628-9_29 · Cited by 2
AceNet: Attention-Guided Context Enhancement for Imbalanced Action Recognition via RF Signals
Sheng, B., Liu, H., Cai, H., Zuo, Y., Zhou, J., & Xiao, F. (2026) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2026.3666114 · Cited by 2
“Convolu-tionalNeuralNetworks(CNNs)andRecurrentNeuralNetworks (RNNs),particularlyLSTMsandGRUs,demonstratesuperior [25],[26],[27].”
Wearable Sensor Applications: Processing of Egocentric Videos and Inertial Measurement Unit Data
Lu, Y., & Velipasalar, S. (2019) · Springer eBooks · doi:10.1007/978-3-030-16949-7_7 · Cited by 1
Liu, Y., Huang, J., & Yang, T. (2023) · Advances in Economics Management and Political Sciences · doi:10.54254/2754-1169/44/20232233 · Cited by 1
Oishi, N., Birch, P., Lago, P., & Roggen, D. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116367 · Cited by 1
Self-citationHuman Activity Recognition using Attribute-Based Neural Networks and Context InformationPreprint
L\"udtke, S., Rueda, F. M., Ahmed, W., Fink, G. A., & Kirste, T. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.04564 · Cited by 11
“An architecture that combines temporal convolutions and recurrent networks is introduced in [ Ord´o´nez and Roggen, 2016 ] .”
BackgroundMethodNimmi, K., Menon, V. G., Janet, B., & Kumar, A. (2020) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-5225-9643-1.ch010 · Cited by 1
Research Progress of Robotic Technologies and Applications in Smart Pig Farms
Zhou, L., Hao, L., Xiong, Y., Qin, H., Bao, A., & Chen, Z. (2026) · Agriculture · doi:10.3390/agriculture16030334 · Cited by 6
Modeling Accurate Human Activity Recognition for Embedded Devices Using Multi-level Distillation.Preprint
Chen, R., Luo, H., Zhao, F., Meng, X., Xie, Z., & Zhu, Y. (2021) · arXiv (Cornell University) · Cited by 4
Multiscale Bidirectional Input Convolutional and Deep Neural Network for Human Activity Recognition
Qiu, Y., Lin, L., Yang, L., Li, D., Song, R., Xu, G., & Shen, S. (2021) · Wireless Communications and Mobile Computing · doi:10.1155/2021/7374177 · Cited by 1
A multi-scale deep neural network for gait recognition using wearable sensors
Sezavar, A., Atta, R., & Ghanbari, M. (2025) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2025.110837 · Cited by 1
Ahmad, N., Leung, H.-F., & Farnia, F. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3555799 · Cited by 4
“These include ALAE-TAE-CutMix+ [5], Attend and Discriminate [13], and DeepConvLSTM [19].”
InfluentialBarot, V., & Patel, R. P. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-10551-7_17 · Cited by 1
Hayama, T. (2021) · International Journal of Service and Knowledge Management · doi:10.52731/ijskm.v5.i2.621 · Cited by 2
Kim, H., & Lee, D. (2021) · ACM Symposium on Applied Computing · doi:10.1145/3412841.3441924 · Cited by 2
“On the other hand, ConvLSTM, one of the advanced deep learning model mentioned in the related work [22], shows lower accuracy(0.”
BackgroundMultivariate Passenger Flow Forecast Based on ACLB Model
Zheng, L., Qi, C., & Zhao, S. (2022) · Lecture notes in electrical engineering · doi:10.1007/978-981-19-2456-9_12 · Cited by 1
Konak, O. (2022) · doi:10.36227/techrxiv.20532060 · Cited by 1
Investigating Domain-agnostic Performance in Activity Recognition using Accelerometer Data
Hasthanasombat, A., Ghosh, A., Spathis, D., & Mascolo, C. (2022) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3544793.3560398 · Cited by 2
“Using this benchmark it is shown that two current state-of-the-art (SotA) deep models [7, 17] face significant performance degradation in unseen domains, even after correcting for factors such as sensor location, sampling rate,e and measurement units.”
BackgroundMethodInfluentialGeneralizing Sensor-Based Human Activity Recognition With Domain-Specific Ensemble Learning
Pan, J., Zhang, L., Hu, Z., Zhang, X., & Cai, X. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3549881 · Cited by 2
“The recurrent neural network LSTM is added behind CNN [28] layers in order to further explore relative importance Authorized licensed use limited to the terms of the applicable license agreement with IEEE.”
Lim, S., Lee, Y. J., & Kim, S. B. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111464 · Cited by 2
Abdelkawy, H. (2021) · theses.fr (ABES) · doi:10.70675/3ae0f056z1a4az4407z8309zd6153c0e83d7 · Cited by 1
Unseen-aware semi-supervised model for robust human activity recognition
Li, Y., Ouyang, Z., & Li, G. (2025) · Multimedia Systems · doi:10.1007/s00530-025-01939-7 · Cited by 1
“In [41], DeepConvLSTM demonstrates the advantages of a deep architecture based on the combination of convolutional and LSTM recurrent layers to perform activity recognition from wearable sensors.”
Waveform optimization of a two-axis smooth impact drive mechanism actuator
Wang, J., Chen, Y., Peng, Y., Song, X., Zhang, Y., & Zhang, M. (2020) · Journal of Intelligent Material Systems and Structures · doi:10.1177/1045389x20951263 · Cited by 1
“This structure is capable of learning order dependence in sequence prediction problems (Francisco and Daniel, 2016; Gers and Schmidhuber, 2001; Ma et al., 2015; Sundermeyer et al., 2012), which can be used for classifying, processing, and making predictions based on time-series data.”
MethodA Deep Dilated Convolutional Self-attention Model for Multimodal Human Activity Recognition
Wang, S., Xiao, S., Wang, Y., Jiang, H., & Zhang, G. (2022) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr56361.2022.9956723 · Cited by 2
Investigation of Transfer Learning for Electricity Load Forecasting
Laitsos, V. M., Vontzos, G., & Bargiotas, D. (2023) · International Conference on Information, Intelligence, Systems & Applications (IISA) · doi:10.1109/iisa59645.2023.10345954 · Cited by 1
“A hybrid CNN-GRU model [22] is a deep learning architecture that combines convolutional neural networks (CNNs) and gated recurrent units (GRUs).”
MethodFLAP: Federated Learning with Attack and Privacy AwarenessPreprint
Xue, W., Ahmed, N., Masood, R., Hu, W. S., Seneviratne, A. P., & Jha, S. K. (2022) · doi:10.36227/techrxiv.19766023.v1 · Cited by 1
Developing an Automatic Evaluation of Exertion Using a Smart Phone
Keerthana, N. L., & Balakrishnan, S. A. (2023) · International Conference on Advanced Computing and Communication Systems (ICACCS) · doi:10.1109/icaccs57279.2023.10112834 · Cited by 1
“The trained model can be deployed on edge devices for real-time classification [15].”
Pepsi, M. B. B., Kumar, N. S., Jeyashree, S., & Subitcha, M. (2024) · Computing · doi:10.1007/s00607-024-01392-w · Cited by 2
“To capture dynamic behaviour, Ordonez and Roggen et al. [19], developed a novel network architecture called DeepConvLSTM that combines feedforward and recurrent units but outperforms CNN alone.”
AI at the Disco: Low Sample Frequency Human Activity Recognition for Night Club Experiences
Gill, A. S., Cabrero, S., Cesar, P. S., & Shamma, D. A. (2020) · International Workshop on Human-centric Multimedia Analysis (HCMA) · doi:10.1145/3422852.3423485 · Cited by 2
“Modern CNNs allow for automatic feature learning from data [1, 13, 22, 27, 33, 35], and have outperformed previous handcrafted feature based methods.”
MethodChen, Y., Yang, L., Yang, J., Li, Y., & Zhou, J. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791432 · Cited by 1
“The hybrid architecture [44], utilizing CNNs to extract local temporal motifs and BiLSTMs to model temporal context [9, 37], is specifically selected to address this multimodal temporal complexity.”
Zhang, L., Zhang, X., Zhang, X., Yu, C., & Liu, X. (2025) · Brain Sciences · doi:10.3390/brainsci15111167 · Cited by 1
“…and long short-term memory units, aiming to simultaneously extract local spatial features and long-term temporal dependencies from EEG signals [33]; (7) TSception: A neural network architecture that uses multi-scale spatiotemporal convolutions, performing exceptionally well in EEG emotion…”
Motion Monitoring for Limb Exercise
Li, Z., Ma, X., & Liu, M. (2019) · International Conference on Information Technology: IoT and Smart City (ICIT) · doi:10.1145/3377170.3377224 · Cited by 1
“[17] combined CNN with LSTM, which was initially designed to process time series signals, to recognize human activities.”
BackgroundDehghani-Sanij, A. A., & Martínez-Hernández, U. (2022) · Springer eBooks · doi:10.1007/978-3-031-07555-1_6 · Cited by 2
Houssein, E. H., Mahdy, M. A., Kayed, M., & Mohamed, W. M. (2025) · Cluster Computing · doi:10.1007/s10586-025-05340-x · Cited by 1
Teng, T. C. I., Wan, J., & Zhang, X. (2024) · Knowledge and Information Systems · doi:10.1007/s10115-024-02154-y · Cited by 3
“Recurrent neural networks (RNN) and their variants [24, 25] are also widely used for processing time-series data to capture dynamic information over time.”
InfluentialGhorrati, Z., & Matson, E. T. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-24209-1_8 · Cited by 1
Trajectory Privacy-Protection Mechanism Based on Multidimensional Spatial–Temporal Prediction
Xi, J., Shi, M., Zhang, W., Xu, Z., & Liu, Y. (2024) · Symmetry · doi:10.3390/sym16091248 · Cited by 1
Alphonse, S. A., Qing, H., & Wang, P. (2025) · Journal of Railway Science and Technology · doi:10.1016/j.jrst.2025.12.002 · Cited by 8
Sensor-Based Personal Activity Recognition Using Mixed 5-Layer CNN-LSTM and Hyperparameter Tunning
Sangisetti, B. R., & Pabboju, S. (2023) · Algorithms for intelligent systems · doi:10.1007/978-981-99-4577-1_2 · Cited by 1
DeepContext: Mobile Context Modeling and Prediction Via HMMs and Deep Learning
Sadhu, V., Zonouz, S., & Pompili, D. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3200947 · Cited by 1
“Ordo-nez et al. [22] propose a technique for human activity recognition based on convolutional and recurrent neural networks, using data collected from wearable devices.”
BackgroundSensor Fusion: Gated Recurrent Fusion to Learn Driving Behavior from Temporal Multimodal DataPreprint
Narayanan, A., Siravuru, A., & Dariush, B. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1910.00628 · Cited by 4
“Behavior Prediction and Image Captioning [14], Activity Recognition[18] and Audiovisual Speech Recognition [13], [19], Speaker Identification [11], Action Classification [10], etc.”
Pellatt, L., & Roggen, D. (2022) · Frontiers in Computer Science · doi:10.3389/fcomp.2022.914330 · Cited by 2
“Although we find that our NAS method can generate better architectures than a random search in considerably less time, and that both produce networks which perform better than our implementations of DeepConvLSTM (Ordóñez and Roggen, 2016) and Attend and Discriminate (Abedin et al.”
BackgroundMethodResultInfluentialSelf-citationA Review of Universal Design in Ambient Intelligence Environments
Bai, A., Mork, H. C., Halbach, T., Fuglerud, K. S., Leister, W., & Schulz, T. W. (2016) · The Internet of Things · Cited by 4
UniMTS: Unified Pre-training for Motion Time Series
Zhang, X., Teng, D., Chowdhury, R. R., Li, S., Hong, D., Gupta, R. K., & Shang, J. (2024) · arXiv · Cited by 9
“We aim for a more generic yet challenging generalization scenario where pre-training and downstream datasets share different label names.”
Wu, X. W., & Zhang, W. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-3055-7_6
Quantization Loss Re-Learning MethodPreprint
Li, K. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.13568 · Cited by 1
Khodabandelou, G., Chibani, A., & Amirat, Y. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11247147
“CNNs learn spatial patterns from sensor streams [1], [2], LSTMs capture long-range temporal context [13], and attention layers sharpen both accuracy and interpretability [14].”
Contrastive Learning Based Human Activity Recognition Using Body Sensors
Xie, C., Zhou, Y., Yu, D., Sun, S., & Zhang, X. (2023) · International Conference on Big Data Computing and Communications (BigCom) · doi:10.1109/bigcom61073.2023.00043 · Cited by 1
“Ord´o˜nez et al. [16] proposed DeepConvLSTM, a general deep activity recognition framework based on convolutional neural network and recurrent neural network, which extracts local features of sensor data and then fed that feature into a recurrent neural network to accomplish human activity…”
BackgroundMekruksavanich, S., & Jitpattanakul, A. (2026) · Joint International Conference on Digital Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications Engineering (ECTI DAMT & NCON) · doi:10.1109/ectidamtncon67592.2026.11460035
Dirchaoui, M., & Abarda, A. (2024) · International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT) · doi:10.1109/3ict64318.2024.10824529
“The CNN model is composed of convolutional layers, pooling layers, and fully connected layers [26], [27], which are arranged in a stacked configuration to facilitate the automatic extraction of features from images [28].”
Li, H.-B., Tian, Y.-H., Gui-wen, W. A. N. G., He, Z.-B., Shao, L.-B., & Lai, J. (2026) · Petroleum Science · doi:10.1016/j.petsci.2026.05.021
Deep Learning Based Breath-by-Breath Tidal Volume Estimation
Ba, M., Pianosi, P. T., & Rajamani, R. (2025) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2025.12.238
Transformer Based Driving Behavior Safety Prediction for New Energy Vehicles
Lin, H., & Yao, J. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-46661-8_43
SPECTRA: An Efficient Spectral-Informed Neural Network for Sensor-Based Activity Recognition
Gurung, D., Ray, L. S. S., Liu, M., Zhou, B., & Lukowicz, P. (2026) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom67906.2026.11524547 · Cited by 2
Yin, D., Xiao, K., Zou, C., Lu, Z., Jiao, C., Chen, M., Rong, H., Xu, Y., & Yang, P. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5340781
A Method of Lower Limb Gait Based on Multi-sensor Data Fusion for Rehabilitation Robot
Lv, J., Li, J., Wang, S., Wang, R., Gao, X., Shi, Y., Lv, P., Liu, H., Zhang, P., Luo, D., et al. (2020) · IEEE International Conference on Mechatronics and Automation (ICMA) · doi:10.1109/icma49215.2020.9233699 · Cited by 1
“Literature [4,5] merged multi-sensor data through fuzzy theory; A neural network method used by researchers in the field of artificial intelligence to achieve the establishment of multi-sensor models and data fusion [6,7]; In addition, some researchers adopted an modified DS evidence theory method to fuse multisensor…”
MethodStable Adversarial Adaptation for Wearable Sensors
Sun, Z., & Zhang, H.-H. (2026) · 2026 IEEE 3rd International Conference on Computer Vision and Deep Learning (DLCV) · doi:10.1109/dlcv69906.2026.11635236
“CNNs and residual networks are widely used for sensor-based activity recognition, while transformers show promise on large-scale time-series tasks [16], [15].”
Sun, T., Zhang, C., Meng, D., Lv, X., & Zhao, Y. (2026) · Measurement · doi:10.1016/j.measurement.2026.122635
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.