Deep Recurrent Neural Networks for Human Activity Recognition
Open accessThe DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index; 5 of the 23 sources its claims cite were checked for support.
Abstract
Adopting deep learning methods for human activity recognition has been effective in extracting discriminative features from raw input sequences acquired from body-worn sensors. Although human movements are encoded in a sequence of successive samples in time, typical machine learning methods perform recognition tasks without exploiting the temporal correlations between input data samples. Convolutional neural networks (CNNs) address this issue by using convolutions across a one-dimensional temporal sequence to capture dependencies among input data. However, the size of convolutional kernels restricts the captured range of dependencies between data samples. As a result, typical models are unadaptable to a wide range of activity-recognition configurations and require fixed-length input windows. In this paper, we propose the use of deep recurrent neural networks (DRNNs) for building recognition models that are capable of capturing long-range dependencies in variable-length input sequences. We present unidirectional, bidirectional, and cascaded architectures based on long short-term memory (LSTM) DRNNs and evaluate their effectiveness on miscellaneous benchmark datasets. Experimental results show that our proposed models outperform methods employing conventional machine learning, such as support vector machine (SVM) and k-nearest neighbors (KNN). Additionally, the proposed models yield better performance than other deep learning techniques, such as deep believe networks (DBNs) and CNNs.
Overview
CoChat Summary
TL;DR
The paper proposes LSTM-based deep recurrent neural network (DRNN) architectures—unidirectional, bidirectional, and cascaded—for human activity recognition from wearable sensor sequences, including variable-length windows. Across benchmark datasets, the proposed models reportedly outperform conventional machine learning and other deep learning baselines.
Key findings
- On UCI-HAD, the unidirectional DRNN with four layers achieved 96.7% overall accuracy, outperforming CNNs, SVM, and sequential ELM.
- On Skoda, the cascaded DRNN with one bidirectional layer and two upper unidirectional layers achieved 92.6% overall accuracy, outperforming HMMs, DBNs, and CNNs.
- On Opportunity, the bidirectional DRNN with three layers outperformed DBNs, SVM, CNNs, and a CNN + unidirectional RNN state-of-the-art method, using F1 score because the dataset is imbalanced.
- On Daphnet FOG, the cascaded DRNN outperformed KNN and CNNs in F1 score; on USC-HAD, the unidirectional four-layer model performed best versus CNNs, LS-SVM, and random forest, though no exact accuracy is given in the text.
Methods
- Three LSTM-based DRNN architectures were developed: unidirectional, bidirectional, and cascaded (bidirectional first layer plus unidirectional upper layers), with late fusion using the sum rule over per-timestep scores.
- Models were evaluated on five public HAR datasets: UCI-HAD, USC-HAD, Opportunity, Daphnet FOG, and Skoda; training used 80% of data and testing used 20%, raw data without further preprocessing, Adam optimization, mean cross entropy, dropout, and a GPU-based TensorFlow framework.
- Performance metrics were precision, recall, accuracy, and F1-score.
Limitations
- Stated: future work includes large-scale and complex human activities, transfer learning between diverse datasets, and resource-efficient DRNN implementation for low-power devices.
- Stated: USC-HAD elevator up/down classes were combined because the model could not differentiate them using only a single IMU; barometer readings would be needed. Daphnet FOG had low recall and precision for the “Freeze” class due to dominance of the “Normal” class, and Opportunity is imbalanced due to the Null class.
- Evident: several dataset-specific results are reported only through figures/confusion matrices without exact accuracy or F1 values in the text, and the paper inconsistently refers to four datasets in the Discussion/Conclusions while the experimental setup describes five.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures












Verification
Claims the paper makes
3 Valid · 0 Not valid · 10 Not checked · 27 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The proposed method outperforms other methods, such as those based on deep believe networks (DBNs) [10], SVM [10], and CNNs [13].”
§ Deep Recurrent Neural Networks for Human Activity Recognition > 6. Results > 6.3. Opportunity
What the sources say
- Not checked yetcites [10] Yang, J.B., Nguyen, M.N., San, P.P., Li, X.L., and Krishnaswamy, S. (2015, January 25–31). Deep convolutional neural networks on multichannel time series for human activity recognition. Proceedings of the 24th International Joint Conference on Artificial Intelligence (IJCAI), Buenos Aires, Argentina.
- Partly supportedcites [13] Ordóñez, F.J., and Roggen, D. (2016). Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors, 16. sourceExperimental — accuracy not yet measured
Our copy supports the SVM and CNN parts: Table 4 (§5.1) shows DeepConvLSTM beating the OPPORTUNITY challenge submissions including SStar (described in Table 3 as a support-vector-machine submission), and §5.1 states it improves by 6% over the CNN results previously reported by Yang et al. and by 5% on average over the baseline CNN. However, the paper makes no head-to-head comparison with deep belief networks on OPPORTUNITY: DBNs appear only in the State of the Art (§2.3) as related work whose feature-extraction applicability is credited to references [31] and [32], with no DBN entry in Tables 3 or 4. The DBN half of the attributed claim is therefore not supported by this source. (LLM, conf=0.80)
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.
Every source the paper cites for this claim was checked and supports it.
In the paper
“It also outperformed the state-of-the-art method, which is a combination of CNNS and unidirectional RNNs [13], for the opportunity dataset.”
§ Deep Recurrent Neural Networks for Human Activity Recognition > 6. Results > 6.3. Opportunity
What the sources say
- Supportedcites [13] Ordóñez, F.J., and Roggen, D. (2016). Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors, 16. source
The citation is attached to the characterisation of the state-of-the-art method on OPPORTUNITY as a combination of CNNs and unidirectional RNNs, and our copy of Ordóñez & Roggen (2016) is exactly that method: Section 3.1 (Architecture) describes DeepConvLSTM as four convolutional layers feeding two stacked recurrent LSTM layers, C(64)-C(64)-C(64)-C(64)-R(128)-R(128)-Sm, with the recurrent layers processing the time axis forward only (no bidirectional component). Section 5.1 (Performance Comparison) reports it outperforming the best OPPORTUNITY-challenge submissions by 6% on average (>9% on gesture recognition without the Null class) and Yang et al.'s CNN by 6%, i.e. it is the top-reported result on that dataset. So the source both is the CNN+unidirectional-RNN architecture and holds the state of the art on OPPORTUNITY that the citing method claims to beat. (LLM, conf=0.86)
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.
- Supported
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We demonstrate the effectiveness of using unidirectional and bidirectional DRNNs for HAR tasks without any additional data-preprocessing or merging with other deep learning methods.”
§ Deep Recurrent Neural Networks for Human Activity Recognition > 1. Introduction
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.
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
“For the UCI-HAD dataset, we found that the unidirectional DRNN model with four layers yields best performance results in terms of per-class precision and recall, as shown in Figure 8a.”
§ Deep Recurrent Neural Networks for Human Activity Recognition > 6. Results > 6.1. UCI-HAD
plainstates an association- 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.
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 proposed method results in an overall accuracy of 92.6%, outperforming other methods such as HMMs [23], DBNs [6], and CNNs [11], as shown in Figure 12b.”
§ Deep Recurrent Neural Networks for Human Activity Recognition > 6. Results > 6.5. Skoda
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: passes — The paper gives a reason why it holds.
23 more claims
Create a free account to see all 40 claims and to verify the 18 references still open.
- Probably not a substantial claim. 4
- It may state two separate claims as one. 7
- The quote only partly states this claim. 2
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 source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“We have applied this network because RvNN can be used to learn distributed, structured data. Therefore, it is ideal for our proposed multimodal IoT-based locomotion prediction system.”
From A Multimodal IoT-Based Locomotion Classification System Using Features Engineering and Recursive Neural Network (2023) · cites this paper as [32]
Experimental — accuracy not yet measuredIn the citing paper
“Papers [33,34] also compared deep-learning methods for HAR, and verified that variants of the LSTM architecture lead to flexible and robust results in several sensor-based HAR datasets.”
From Sensor Data Acquisition and Multimodal Sensor Fusion for Human Activity Recognition Using Deep Learning (2019) · cites this paper as [33]
In the citing paper
“Previous studies have shown that LSTMs have high performance in wearable HAR [199,202,203].”
From Deep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances (2022) · cites this paper as [202]
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 minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Deep Recurrent Neural Networks for Human Activity Recognition > 8. Conclusions
“Additionally, we empirically evaluated our models by conducting experiments on four miscellaneous benchmark datasets.”
§ Deep Recurrent Neural Networks for Human Activity Recognition > 5. Experimental Setup > 5.1. Datasets of Human Activities
“In order to train and evaluate the proposed models, we considered five public benchmark datasets for HAR.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Deep Recurrent Neural Networks for Human Activity Recognition > 5. Experimental Setup > 5.3. Performance Metrics
“Overall Accuraccy=TP+TNTP+TN+FP+FN, where TP=∑c=1Ctpc is the overall true positive rate for a classifier on all classes, TN=∑c=1Ctnc is the overall true negative rate, FP =∑c=1Cfpc is the overall false positive rate, and FN=∑c=1Cfnc is the overall false negative rate.”
§ Deep Recurrent Neural Networks for Human Activity Recognition > 5. Experimental Setup > 5.3. Performance Metrics
“Accuracy: Measures the proportion of correctly predicted labels over all predictions:”
Citations
Cites
Bidirectional recurrent neural networks
Schuster, M., & Paliwal, K. K. (1997) · IEEE Transactions on Signal Processing · doi:10.1109/78.650093 · Cited by 10,172 · cited 1×
“It includes two parallel LSTM tracks: forward and backward loops for exploiting context from the past and future of a specific time step in order to predict its label [17].”
BackgroundTensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., et al. (2016) · arXiv · doi:10.48550/arxiv.1603.04467 · Cited by 9,408 · cited 1×
“Training was conducted on a GPU-based TensorFlow framework in order to utilize the parallel computation power of a GPU [25].”
MethodSpeech 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 3×
“Using sufficient number of DRNN layers can result in a very powerful model for transforming raw data into a more abstract representation, as well as for learning the temporal dependencies in time series data [1].”
BackgroundMethodA systematic analysis of performance measures for classification tasks
Sokolova, M., & Lapalme, G. (2009) · Information Processing & Management · doi:10.1016/j.ipm.2009.03.002 · Cited by 7,515
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Wu, Y., Schuster, M., Chen, Z., Le, Q. V., Norouzi, M., Macherey, W., Krikun, M., Cao, Y., Gao, Q., Macherey, K., et al. (2016) · arXiv · doi:10.48550/arxiv.1609.08144 · Cited by 5,319 · cited 1×
“The third model architecture, shown in Figure 6, is motivated by [18].”
BackgroundSupervised Sequence Labelling with Recurrent Neural Networks
Graves, A. (2012) · Studies in Computational Intelligence · doi:10.1007/978-3-642-24797-2 · Cited by 3,120
Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Ordóñez, F., & Roggen, D. (2016) · Sensors · doi:10.3390/s16010115 · Cited by 2,761
Phoneme recognition using time-delay neural networks
Waibel, A., Hanazawa, T., Hinton, G., Shikano, K., & Lang, K. J. (1989) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/29.21701 · Cited by 2,643 · cited 1×
“To capture local dependencies, CNNs use parameter sharing across time—applying the same convolutional kernel at each time segment—and local connectivity—neurons receiving inputs from small groups of input samples—between adjacent layers [12].”
MethodLSTM neural networks for language modeling
Sundermeyer, M., Schlüter, R., & Ney, H. (2012) · Interspeech 2012 · doi:10.21437/interspeech.2012-65 · Cited by 1,963 · cited 1×
“They have achieved promising performance enhancements in many technical applications, such as speech recognition [1], language modeling [2], video processing [3], and many other sequence labeling tasks [4].”
Background(2013) · RECERCAT (Consorci de Serveis Universitaris de Catalunya) · Cited by 1,959
Research works printed but not linked: 1
- Goodfellow, I., Bengio, Y., and Courville, A. (2016). Optimization for Training Deep Models. Deep Learning, The MIT Press.
Cited by
Items 301–400 of 545
Sensor-Based Automatic Recognition of Construction Worker Activities Using Deep Learning Network
Tezcan, Ö., Akçay, C., Sari, M., & Cavus, M. (2025) · Sensors · doi:10.3390/s25133988 · Cited by 6
“…although recent DL-based HAR systems using public datasets have reported high accuracies—e.g., 98% using LSTM on Parkinson’s data [15], and 93–95% with CNN or hybrid models [60,61]—our results match or exceed these despite being based on more variable and noisy real-world construction data.”
Efficient few-shot human activity recognition via meta-learning and data augmentation
Vettoruzzo, A., Bouguelia, M.-R., & Rognvaldsson, T. S. (2025) · Neural Computing and Applications · doi:10.1007/s00521-025-11408-x · Cited by 9
Deep Learning Algorithms for Human Activity Recognition in Manual Material Handling Tasks
Bassani, G., Avizzano, C. A., & Filippeschi, A. (2025) · Sensors · doi:10.3390/s25216705 · Cited by 7
“Murad and Pyun [41] proposed one of the first uses of LSTM networks for HAR.”
Human Activity Recognition Using Convolutional Neural Networks
Awad, O. F., Ahmed, S. R., Shaker, A. S., Majeed, D. A., Hussain, A.-S. T., & Taha, T. A. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-62871-9_20 · Cited by 6
Gonçalves, H. R., Martins, L., Ribeiro, N. F., Abreu, L., Rodrigues, A. M., & Santos, C. P. (2025) · Engineering applications of artificial intelligence · doi:10.1016/j.engappai.2025.110706 · Cited by 6
McDaniel, C., & Quinn, S. P. (2018) · Proceedings of the Python in Science Conferences · doi:10.25080/majora-4af1f417-005 · Cited by 7
Context-Aware Human Activity Recognition (CA-HAR) Using Smartphone Built-In Sensors
Fan, L., Haghighi, P. D., Zhang, Y., Forkan, A. R. M., & Jayaraman, P. P. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20436-4_11 · Cited by 6
A Framework for Human Activity Recognition Application for Therapeutic Purposes
Hamido, M., Mosallam, K., Diab, O., Amin, D., & Atia, A. (2023) · Intelligent Methods, Systems, and Applications (IMSA) · doi:10.1109/imsa58542.2023.10217417 · Cited by 6
“Although RNNs seem promising, concerns have been raised regarding an issue in which it cannot continue to learn, often referred to as the Vanishing Gradient Problem [13].”
BackgroundInterrogation, Intelligence and the Issue of Human Rights
Newbery, S., Brecher, B., Sands, P., & Stewart, B. (2009) · Intelligence & National Security · doi:10.1080/02684520903209381 · Cited by 6
Deep 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
Neuroevolution based hierarchical activation function for long short-term model network
Vijayaprabakaran, K., & Sathiyamurthy, K. (2021) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-020-02889-w · Cited by 6
Li, D. (2021) · Computational Intelligence and Neuroscience · doi:10.1155/2021/2545151 · Cited by 7
RetractedDeep Incremental Learning for Personalized Human Activity Recognition on Edge Devices
Younan, S., & Abu-Elkheir, M. (2022) · Canadian Journal of Electrical and Computer Engineering · doi:10.1109/icjece.2022.3199227 · Cited by 7
“…annotation, and analysis, a machine or deep learning algorithm can be trained on benchmarked human activity data that are properly annotated, and the resulting model can be serialized and deployed in the human activity recognition (HAR) component within a given application [2], [3], [4], [5].”
BackgroundPosition-Aware Indoor Human Activity Recognition Using Multisensors Embedded in Smartphones
Wang, X., Wang, Y., & Wu, J. (2024) · Sensors · doi:10.3390/s24113367 · Cited by 6
“Murad et al. [17] used naive Bayesian and K-nearest neighbor algorithms to explore the ability of integrating built-in accelerometers and gyroscopes in smartphones to recognize human body activities.”
MethodMOMOWO: Women Designers, Craftswomen, Architects and Engineers between 1918 and 1945
Groot, M., Seražin, H., Franchini, C., & Garda, E. M. (2017) · Womenʼs creativity · doi:10.3986/wocrea/1/momowo1 · Cited by 6
An Activity Recognition Application Based on Markov Decision Process Through Fish Eye Camera
Öner, A., Albayrak, M. S., Furkan, G., & Atakli, I. M. (2019) · IEEE International Conference on Artificial Intelligence and Knowledge Engineering (AIKE) · doi:10.1109/aike.2019.00050 · Cited by 6
“[28] have proposed an innovative human activity recognition method based on long-short term memory”
MethodZheng, L., Li, S., Zhu, C., & Gao, Y. (2019) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3341162.3344851 · Cited by 6
“On the other hand, considering the periodicity of the signal obtained by smartphone sensors, FFT spectrums of a short-time window was computed in [1, 3, 5, 10] to analyze the frequency characteristics of smartphone sensor signals as the preprocessed feature.”
MethodNguyen, H. D., Tran, K. P., Zeng, X., Koehl, L., & Tartare, G. (2020) · Springer series in reliability engineering · doi:10.1007/978-3-030-43412-0_13 · Cited by 7
Hybrid deep neural networks to infer state models of black-box systems
Mashhadi, M. J., & Hemmati, H. (2020) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1145/3324884.3416559 · Cited by 9
“vity that human was performing in different time intervals. The sensors can be body worn accelerometers, or more generic sensors such as the ones found in a smart watch or a smartphone. Murad et al., [54] have shown deep RNNs outperform fully convolutional networks and deep belief networks in HAR task. Hybrid models a…”
BackgroundBubblenet: A Disperse Recurrent Structure To Recognize Activities
Bastos, I. L. O., C., M. V. H., & Schwartz, W. R. (2020) · Proceedings - International Conference on Image Processing · doi:10.1109/icip40778.2020.9190769 · Cited by 6
“D convolutions [5] and recurrent neural networks [10, 11].”
BackgroundMethodGarcía-Domínguez, A., Galvan-Tejada, C. E., Zanella-Calzada, L. A., Gamboa-Rosales, H., Galván-Tejada, J. I., Padilla, J. M. C., Luna-García, H., Arceo-Olague, J. G., & Magallanes‐Quintanar, R. (2020) · PeerJ Computer Science · doi:10.7717/peerj-cs.308 · Cited by 6
Chen, Y. (2024) · Journal of Electrical Systems · doi:10.52783/jes.1387 · Cited by 5
“In [12], DRNNs for human activity recognition is constructed, emphasizing their capability to grasp dependencies across extensive ranges in input sequences of varying lengths.”
BackgroundKhatiwada, P., Chatterjee, A., & Subedi, M. (2021) · IEEE International Conference on High Performance Computing and Communications; IEEE International Conference on Smart City; IEEE International Conference on Data Science and Systems (HPCC/SmartCity/DSS) · doi:10.1109/hpcc-dss-smartcity-dependsys53884.2021.00187 · Cited by 5
Segment-Based Unsupervised Learning Method in Sensor-Based Human Activity Recognition
Takenaka, K., Kondo, K., & Hasegawa, T. (2023) · Sensors · doi:10.3390/s23208449 · Cited by 5
To Build and Evaluate Irregular Human Activity Detection System Using Deep CNN-LSTM Methodology
Pandey, V. P., Gulvi, T., Waghmare, S. R., Dalvi, P., Kesarkar, R., & Dayal, D. (2024) · IEEE International Conference on Advances in Computing Research on Science Engineering and Technology (ACROSET) · doi:10.1109/acroset62108.2024.10743555 · Cited by 5
Applying MLP-Mixer and gMLP to Human Activity Recognition
Miyoshi, T., Koshino, M., & Nambo, H. (2025) · Sensors · doi:10.3390/s25020311 · Cited by 6
“Murad et al. [16] introduced a model using Long Short-Term Memory (LSTM), while Li et al. [28] developed a Bidirectional LSTM approach.”
Zhao, J., Zeng, P., Chen, C., Dong, Z., & Han, J. (2021) · Intelligent Automation & Soft Computing · doi:10.32604/iasc.2021.016966 · Cited by 5
“In practice, LSTM is used to solve the problem of the explosion and disappearance of the RNN gradient [19].”
MethodA customized 1D-CNN approach for sensor-based human activity recognition
(2022) · International Journal of Advanced Technology and Engineering Exploration · doi:10.19101/ijatee.2021.874828 · Cited by 7
Omae, Y., Sakai, Y., & Takahashi, H. (2023) · AIMS Mathematics · doi:10.3934/math.2024041 · Cited by 5
Pîrjan, A., Căruțașu, G., & Petroșanu, D.-M. (2018) · Energies · doi:10.3390/en11102623 · Cited by 5
Framework of Sequence Chunking for Human Activity Recognition Using Wearables
Zhang, W., Qin, L., Zhong, W., Guo, X., & Wang, G. (2019) · International Conference on Image, Video and Signal Processing (IVSP) · doi:10.1145/3317640.3317647 · Cited by 5
“Also, recurrent models [7,8,9] were adopted to model temporal dependencies.”
MethodRobust Sensor-based Human Activity Recognition with Snippet Consensus Neural Networks
Huang, Y., Lee, M.-C., Tseng, V. S. M., Hsiao, C.-J., & Huang, C.-C. (2019) · Wearable and Implantable Body Sensor Networks · doi:10.1109/bsn.2019.8771073 · Cited by 5
“[10] proposed a deep RNN that are able to capture the long-range dependencies in different length of input signal data.”
BackgroundDeep Recurrent Neural Networks for Human Activity Recognition During Skiing
Pawlyta, M., Hermansa, M., Szczęsna, A., Janiak, M., & Wojciechowski, K. W. (2019) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-31964-9_13 · Cited by 5
Human Activity Recognition Using Positioning Sensor and Deep Learning Technique
Mohite, A., Rege, P. P., & Chakravarty, D. (2021) · Lecture notes in electrical engineering · doi:10.1007/978-981-15-8391-9_34 · Cited by 5
Deep Bidirectional GRU network for human activity recognition using Wearable inertial sensors
Zhao, S., Wei, H., & Zhang, K. (2022) · International Conference on Electronic Communication and Artificial Intelligence (ICECAI) · doi:10.1109/iwecai55315.2022.00054 · Cited by 5
Transfer Learning Model Knowledge Across Multi-Sensors Locations Over Body Sensor Network
Qian, X., Chen, H., Cai, Y. M., Chu, K., Xu, W., & Huang, M.-C. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3166187 · Cited by 6
“To better evaluate and assess the performance of the multi-classification in the experiments, the metric of overall accuracy, Macro-precision, Macro-recall, and Macro-F1 [47], [48] is used.”
MethodKuschan, J., Filaretov, H., & Krüger, J. (2022) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin51773.2022.9976078 · Cited by 5
“…and Uni-directional LSTM-based Deep Recurrent Neural Network (C-DRNN) [14], Deep Convolutional Long Short-Term Memory (DC-LSTM) [15] and CNN-2 [16]. a) C-DRNN: The Cascaded Bidirectional and Unidirec-tional Long Short-Term Memory (LSTM)-based Deep Recurrent Neural Network (DRNN) from [16] is…”
BackgroundAshwin, M., Jagadeesan, D., Kumar, M. R., Murugavalli, S., Krishna, A. C., & Ammisetty, V. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-19095-x · Cited by 7
“Murad et al., [12] have utilized Deep Recurrent Neural Networks (DRNNs), be able to detect extended-choice dependency for differential input data.”
A single feature for human activity recognition using two-dimensional acoustic array
Guo, X., Su, R., Hu, C., Xiao-dong, Y., Wu, H., & Nakamura, K. (2019) · Applied Physics Letters · doi:10.1063/1.5096572 · Cited by 5
Sun, P., Han, H., Sun, Y.-K., Wang, X., Liu, X., Zhou, B.-Y., Wang, L.-F., Zhang, Y.-Q., Pan, Z.-G., Huang, B.-J., et al. (2024) · ULTRASONOGRAPHY · doi:10.14366/usg.24172 · Cited by 5
Activity Recognition using Multi-Class Classification inside an Educational Building
Das, A., & Kjærgaard, M. B. (2020) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops48775.2020.9156269 · Cited by 4
“AR also has many notable and compelling applications ranging from healthcare to fitness, elderly assistance, ambient sensing, indoor navigation, gaming, are to name a few [2], [3], [6], [7].”
BackgroundPersonalized Human Activity Recognition using Hypergraph Learning with Fusion Features
Wang, L., Sun, J., Pan, T., Ye, Y., He, W., & Yang, K. (2021) · International Conference on Electronics Technology (ICET) · doi:10.1109/icet51757.2021.9451065 · Cited by 4
“We present the results of some existing methods of HAR on the public dataset USC-HAD [19],[20],[21] in TABLE I.”
MethodChuvashov, R. D., Baranova, A. A., Khokhlov, K. O., & Verbitskiy, E. V. (2020) · International Congress on Energy Fluxes and Radiation Effects (EFRE) · doi:10.1109/efre47760.2020.9242148 · Cited by 4
“Well-developed pattern recognition methods for multi-time-series (multi-TS) identification [8–11] require infrastructure: the dataset representing optical attenuation of fluorophores in contact with analytes is needed, as well as the device capable of registering and transferring that data.”
MethodFataniya, A. G., & Modi, H. P. (2022) · International Journal of Computing and Digital Systems · doi:10.12785/ijcds/120188 · Cited by 4
“DRNN for HAR (2017) [11] DRNN Sensor UCI-HAD 6 96.”
Yusup, N., Khan, A. S., Sarbini, I. N., & Mohamad, N. Z. (2022) · International Conference on Digital Transformation and Intelligence (ICDI) · doi:10.1109/icdi57181.2022.10007121 · Cited by 4
“LSTM-DRNN Model [16] After that, stochastic crossover is carried out between the three fundamental moves in order to locate the most recent binary grey wolf location.”
BackgroundImproving EEG-Based Epileptic Seizures Detection Using Self-Normalizing Neural Network
Mekruksavanich, S., Jantawong, P., & Jitpattanakul, A. (2023) · International Conference on Big Data Analytics and Practices (IBDAP) · doi:10.1109/ibdap58581.2023.10271979 · Cited by 4
“However, in conventional machine learning techniques, the selection of features and classifiers often involves a trial-and-error process [6].”
BackgroundArabacı, H., Uçar, K., & Çimen, H. (2024) · Electrical Engineering · doi:10.1007/s00202-024-02392-x · Cited by 6
“In time series analysis, Recurrent Neural Network (RNN) architecture was developed to capture the correlation between data points [32, 33].”
Multi-attention deep recurrent neural network for nursing action evaluation using wearable sensor
Zhong, Z., Lin, C., Ogata, T., & Ota, J. (2020) · ACM International Conference on Intelligent User Interfaces (IUI) · doi:10.1145/3377325.3377530 · Cited by 4
“Thereafter, owing to the better performance of recurrent neural networks (RNNs) in the time-series domain, several variants of RNNs, such as multicolumnBiRNN [29], DeepConvLSTM [23], ResBidirLSTM [34], and DRNN [21] have been proposed, which represent the state-of-theart level wearable sensor-based HAR.”
BackgroundMethodCoelho, Y., Nguyen, B., Santos, F. A., Krishnan, S., & Bastos-Filho, T. (2022) · World Congress on Medical Physics and Biomedical Engineering, September 7 - 12, 2009, Munich, Germany · doi:10.1007/978-3-030-70601-2_276 · Cited by 6
Raza, A. (2023) · theses.fr (ABES) · doi:10.70675/ac35dc8az89ccz4977z8f25zb7c8937f7cf0 · Cited by 22
Marhraoui, Y. E., Bouilland, S., Deseur, M., Boukallel, M., Bouchigny, S., Panëels, S., Anastassova, M., Muhla, F., Monsaingeon, J., Sauquet, D., et al. (2025) · IRBM · doi:10.1016/j.irbm.2025.100907 · Cited by 4
Akbari, S., Horemans, H. L. D., Bussmann, J. B. J., & Zgonnikov, A. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.111167 · Cited by 3
Nasir, I., Khan, M. H., & Farid, M. S. (2025) · Journal of the Franklin Institute · doi:10.1016/j.jfranklin.2025.108218 · Cited by 3
Deep Learning-Based Human Activity Recognition Using Binary Ambient Sensors
Zhao, Q., Ghasemi, A., Saif, A., & Bossard, L. (2026) · Electronics · doi:10.3390/electronics15020428 · Cited by 3
Patrick, B., Kanjo, E., & Kaiwartya, O. (2026) · Smart Agricultural Technology · doi:10.1016/j.atech.2026.101986 · Cited by 5
Behavior recognition research based on reinforcement learning for dynamic key feature selection
Zhang, T., Ma, C., Sun, H., Yan, L., Wang, B., & Fang, Y. (2022) · International Symposium on Advances in Informatics, Electronics and Education (ISAIEE) · doi:10.1109/isaiee57420.2022.00054 · Cited by 3
“The authors of the literature[9] proposed the use of deep recurrent neural networks (DRNNs) to construct recognition models that are capable of capturing long-range dependencies in variable-length input sequences, and the experimental results outperformed traditional machine learning methods.”
BackgroundAutomatic gate-to-gate time recognition from audio recordings in slalom skiing using neural networks
Menhorn, F., Hummel, C., Huber, A., Waibel, K., Bungartz, H., & Spitzenpfeil, P. (2024) · Current Issues in Sport Science (CISS) · doi:10.36950/2024.3ciss003 · Cited by 3
“Murad & Pyun (2017) present an approach using neural networks to recognize a variety of human activity, using long short-term memory networks.”
BackgroundMethodAlam, M. T., Dasgupta, S., & Roy, K. (2024) · International Conference on Computer and Information Technology (ICCIT) · doi:10.1109/iccit64611.2024.11022452 · Cited by 5
“Convolutional Neural Networks (CNNs) [28] and Recurrent Neural Networks (RNNs) [29] significantly improved HAR.”
An Effective Approach for Wearable Sensor-Based Human Activity Recognition in Elderly Monitoring
Errafik, Y., Dhassi, Y., Baghrous, M., & Kenzi, A. (2025) · BioMedInformatics · doi:10.3390/biomedinformatics5030038 · Cited by 4
Understanding the audience experience of contemporary visual arts at Geevor Mine World Heritage Site
Bagnall, G., & Randall, J. M. (2020) · Routledge eBooks · doi:10.4324/9780429053498-14 · Cited by 3
AR-T: Temporal Relation Embedded Transformer for the Real World Activity Recognition
Kim, H., & Dongman (2022) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-94822-1_40 · Cited by 5
SVM Time Series Classification of Selected Gait Abnormalities
Rostovski, J., Krivošei, A., Kuusik, A., Ahmadov, U., & Alam, M. M. (2022) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-95593-9_16 · Cited by 3
(2017) · BRILL eBooks · doi:10.1163/9789004360846 · Cited by 3
Out-of-distribution in Human Activity Recognition
Roy, D., Komini, V., & Girdzijauskas, Š. (2022) · Swedish Artificial Intelligence Society Workshop (SAIS) · doi:10.1109/sais55783.2022.9833052 · Cited by 3
Musharu, T., & Vogts, D. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-15342-6_12 · Cited by 3
Galindo-López, C. R., Beltrán, J., Pérez, C. B., Macías, A., & Castro, L. A. (2022) · Mexican International Conference on Computer Science · doi:10.1109/enc56672.2022.9882907 · Cited by 4
Rajamoney, J., & Ramachandran, A. (2022) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.7468 · Cited by 3
Analyzing the Performance of Machine Learning Models in Music Genre Classification
Sharma, L. (2023) · International Journal for Research in Applied Science and Engineering Technology · doi:10.22214/ijraset.2023.54344 · Cited by 3
Deep Learning Networks for Complex Activity Recognition Based on Wrist-Worn Sensor
Mekruksavanich, S., & Jitpattanakul, A. (2023) · IEEE Region 10 Conference (TENCON) · doi:10.1109/tencon58879.2023.10322430 · Cited by 3
“Consequently, recurrent neural networks (RNNs) are utilized in HAR to extract temporal information from wearable sensor data, addressing the temporal aspect [16].”
BackgroundImproving Lower Limb Activity Recognition Based on Inertial Sensors Using CNN-LSTM Network
Mekruksavanich, S., Jantawong, P., & Jitpattanakul, A. (2023) · Research, Invention, and Innovation Congress: Innovative Electricals and Electronics (RI2C) · doi:10.1109/ri2c60382.2023.10355936 · Cited by 3
“To address this, LLAR [12] incorporates recurrent neural networks (RNNs) to capture the temporal information derived from wearable sensor data.”
BackgroundAnomalous Human Action Recognition with Deep Learning Technique
Pallewar, M. G., Pawar, V. R., & Gaikwad, A. N. (2024) · International Conference on Computing for Sustainable Global Development · doi:10.23919/indiacom61295.2024.10498200 · Cited by 3
“Comparing with methods used in recent years [19] – [23], the present methodology achieved 97.8% accuracy for six class activities.”
MethodInfluentialYoshida, Y., Miwa, H., Nakae, S., Ogiso, S., Ichikari, R., & Okuma, T. (2025) · IEEE Access · doi:10.1109/access.2025.3550579 · Cited by 3
“Factors such as sensor type and configuration, data collection environment, class count, and measurement duration significantly affect performance [11].”
Efficient Keyset Design for Neural Networks Using Homomorphic Encryption
Joo, Y., Ha, S., Oh, H., & Paek, Y. (2025) · Sensors · doi:10.3390/s25144320 · Cited by 3
Multi-input 1-dimensional deep belief network: action and activity recognition as case study
Nickfarjam, A. M., & Ebrahimpour-Komleh, H. (2019) · Multimedia Tools and Applications · doi:10.1007/s11042-018-7076-0 · Cited by 3
Antenucci, A., Brancati, A., Mazzaro, S., Bastianelli, G., Rovella, R., Massa, A., & Matta, W. (2021) · International Conference on Computer, Control and Robotics (ICCCR) · doi:10.1109/icccr49711.2021.9349276 · Cited by 3
“…to correlate data and images coming from the ePassport module by RFID, the system is able to perform real-time face detection and recognition using modern face recognition pipelines, based on FaceNet [2] approach, assuring a good accuracy on different lighting conditions and face poses [7,8].”
MethodInertial Measurement Unit based Human Action Recognition for Soft-Robotic Exoskeleton
Kuschan, J., Burgdorff, M., Filaretov, H., & Krüger, J. (2021) · IOP Conference Series Materials Science and Engineering · doi:10.1088/1757-899x/1140/1/012020 · Cited by 3
“The F1 scores without Null class performs better for both the default and reduced models of the C-DRNN and reduced models of DC-LSTM and CNN-2 better, than for the Opportunity data set.”
BackgroundMethodInfluentialPRESENCIA Y REALIDAD DE LA PSICOLOGÍA POLÍTICA FRANCESA
Dorna, A. (1998) · Psicología Política · Cited by 7
RestHAR: Residual Feature Learning Transformer for Human Activity Recognition from Multi-sensor Data
Shome, D. (2021) · International Conference on Soft Computing & Machine Intelligence (ISCMI) · doi:10.1109/iscmi53840.2021.9654816 · Cited by 3
“In [9], recurrent architectures such as RNNs and LSTMs were explored for HAR.”
MethodA Hybrid Deep Learning Model for Human Activity Recognition Using Wearable Sensors
Gaurav, K., Roy, B., & Bharti, J. (2022) · Algorithms for intelligent systems · doi:10.1007/978-981-16-9650-3_16 · Cited by 4
Classifying falls using out-of-distribution detection in human activity recognition
Roy, D., Komini, V., & Girdzijauskas, Š. (2023) · AI Communications · doi:10.3233/aic-220205 · Cited by 2
“Popular choices of deep-learning architectures include CNN based on 1D convolution of time-series data [14], recurrent neural-networks [28], autoencoder-based architectures [34].”
Velusamy, A., Akilandeswari, J., Karthikeyan, A. N., Jaleel, D. A., Viswanathan, R. V., & Mohanraj, B. (2024) · OPJU International Technology Conference on Smart Computing for Innovation and Advancement in Industry (OTCON) · doi:10.1109/otcon60325.2024.10687848 · Cited by 2
“Furthermore, the study's dependence on a very small sample size of 10 individuals prompts questions regarding the dataset's suitability for reliable model training and broad suitability [10].”
BackgroundKataoka, S., Oba, M., & Nonaka, H. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110232 · Cited by 3
Incept-LSTM: deep learning model for action recognition
Verma, U., Tyagi, P., & Aneja, M. K. (2025) · Journal of Instrumentation · doi:10.1088/1748-0221/20/10/t10002 · Cited by 2
“A model based on Recurrent Neural Network was proposed in [9] where Long Short Term Memory (LSTM) was utilized, both unidirectionally and bidirectionally, to capture the long term dependencies in the variable length input of data sequences.”
Rep-MMB: Bridging Mobile CNN and Transformer for Sensor-Based Human Activity Recognition
Liu, J., Zhang, L., Liu, X., Chen, G., Fu, Z., Huang, W. R., Wu, H., & Song, A. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3685081 · Cited by 2
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
“CNNs learn discriminative patterns directly from raw signals [9], while RNNs (LSTMs, GRUs) capture temporal dependencies [10].”
Toward Wearable Sensor-Based Human Activity Recognition: A Survey
Zou, H., Chen, Z., Pan, Y., & Li, J. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3702134 · Cited by 3
Framework for Human Activity Recognition on Smartphones and Smartwatches
Mitrevski, B., Petreski, V., Gjoreski, M., & Stojkoska, B. R. (2018) · Communications in computer and information science · doi:10.1007/978-3-030-00825-3_8 · Cited by 2
Human Activity Recognition Using Deep Learning-Based Approach
Rahman, M., & Das, T. (2021) · Lecture notes in networks and systems · doi:10.1007/978-981-16-1089-9_63 · Cited by 2
Human Activity Recognition based on WaveNet
Hao, T. (2021) · IEEE World Forum on Internet of Things (WF-IoT) · doi:10.1109/wf-iot51360.2021.9595629 · Cited by 2
Landmark Detection Based on Human Activity Recognition for Automatic Floor Plan Construction
Huang, Z., Poslad, S., Li, Q., Li, J., & Chen, C. (2022) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-031-24386-8_25 · Cited by 2
Abnormal Activity Recognition in Private Places Using Deep Learning
Suthar, A., Mangrolia, J. R., & Patel, R. (2023) · International Journal of Scientific Research in Computer Science Engineering and Information Technology · doi:10.32628/cseit228688 · Cited by 2
One-D Convolution Neural Network Models for Human Activity Recognition using mHealth Datasets
Yogesh, K. M., Arpitha, S., & Gad, I. (2023) · International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) · doi:10.1145/3616901.3616933 · Cited by 2
“Alternative methodologies [14] have incorporated a combination of CNN or deep RNN with LSTM.”
MethodAI Trainer : Video-Based Squat Analysis
Garande, P. A., Patil, K., Deshmukh, R., Gurav, S., & Yadav, C. (2024) · International Journal of Scientific Research in Science Engineering and Technology · doi:10.32628/ijsrset2411221 · Cited by 2
Human Motion Recognition Information Based on LSTM Recurrent Neural Network Algorithm
Long, Y., & Pan, L. (2024) · IEEE International Conference on Information Systems and Computer Aided Education (ICISCAE) · doi:10.1109/iciscae62304.2024.10761698 · Cited by 2
“Typically employing a softmax activation function, the output layer transforms the output of the fully connected layer into a probability distribution across various motion patterns, thus achieving precise recognition of human motion patterns [8].”
Achieving High-Accuracy Human Activity Recognition Using BERT-Based Classwise Ensemble Models
Fuad, K. M. N. R., Ahmed, M., Abir, M. G. R., Rahman, M. A., Zamil, M. Z. H., Mridha, M. F., & Shin, J. (2024) · IEEE International Conference on Computing, Applications and Systems (COMPAS) · doi:10.1109/compas60761.2024.10796319 · Cited by 2
Analysis of Multimodal Sensor Systems for Identifying Basic Walking Activities
Mitchell, J. K., Dehghani-Sanij, A. A., Xie, S. Q., & O’Connor, R. J. (2025) · Technologies · doi:10.3390/technologies13040152 · Cited by 3
“USC-HAD was chosen because this dataset has been widely explored in the literature since its publication [15,16,34].”
Computers and Independent Study: Student Perspectives
Jarvis, H. (2009) · IGI Global eBooks · doi:10.4018/978-1-59904-994-6.ch023 · Cited by 2
Target-Aware Prediction of Tool Usage in Sequential Repair Tasks
Nabizadeh, N., Heckmann, M., & Kolossa, D. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-64580-9_13 · Cited by 2
Lina, L., Su, J., & Ajienegoro, D. (2021) · Jurnal Muara Sains Teknologi Kedokteran dan Ilmu Kesehatan · doi:10.24912/jmstkik.v5i2.12010 · Cited by 2
Human Activity Recognition Using Stacked LSTM
Benhaili, Z., Kabbaj, I., Balouki, Y., & Moumoun, L. (2022) · Lecture notes in networks and systems · doi:10.1007/978-3-030-91738-8_4 · Cited by 2
Metadata
- Authors
- Murad, A., & Pyun, J.-Y.
- Affiliations
- Chosun University (KR)
- Corresponding author
- Pyun, Jae-Young
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2017-11-06
- Volume
- 17
- Issue
- 11
- Pages
- 2556
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC5712979
- OpenAlex
- W2767979715
- Cited by
- 524
- References
- 33
- Citations by year
- 2026: 27 · 2025: 45 · 2024: 57 · 2023: 60 · 2022: 81 · 2021: 82 · 2020: 80 · 2019: 45 · 2018: 19 · 2017: 2
- FWCI
- 16.57
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- 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 › Computer Vision and Pattern Recognition › Human Pose and Action Recognition
- Keywords
- human activity recognition, deep recurrent neural networks, long short-term memory, wearable sensors, temporal dependencies, bidirectional LSTM, wearable sensor data
- MeSH terms
- Machine Learning, Algorithms, Humans, Human Activities, Neural Networks, Computer
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 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash).
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.