A Comprehensive Study of Activity Recognition Using Accelerometers
Open accessPublished version of a preprint
The DOI matches its Crossref record; 98 of the 99 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
This paper serves as a survey and empirical evaluation of the state-of-the-art in activity recognition methods using accelerometers. The paper is particularly focused on long-term activity recognition in real-world settings. In these environments, data collection is not a trivial matter; thus, there are performance trade-offs between prediction accuracy, which is not the sole system objective, and keeping the maintenance overhead at minimum levels. We examine research that has focused on the selection of activities, the features that are extracted from the accelerometer data, the segmentation of the time-series data, the locations of accelerometers, the selection and configuration trade-offs, the test/retest reliability, and the generalisation performance. Furthermore, we study these questions from an experimental platform and show, somewhat surprisingly, that many disparate experimental configurations yield comparable predictive performance on testing data. Our understanding of these results is that the experimental setup directly and indirectly defines a pathway for context to be delivered to the classifier, and that, in some settings, certain configurations are more optimal than alternatives. We conclude by identifying how the main results of this work can be used in practice, specifically in experimental configurations in challenging experimental conditions.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures














Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Breiman, L. (2001) · Machine Learning · doi:10.1023/a:1010933404324 · Cited by 133,538
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 · cited 1×
“Convolutional Neural Networks Convolutional Neural Networks (CNNs), also referred to as ConvNets, are a widely used extension of MLPs which have seen great success in computer vision tasks [80].”
BackgroundHochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 13×
“Our experiments use LSTM since they have been shown to perform similarly to other methods (e.g., GRU) in many settings [87].”
MethodInfluentialAdam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“All networks were trained with the Adam optimiser [88] and parameters are tuned to minimise categorical cross entropy.”
MethodLearning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 2×
“Our experiments use LSTM since they have been shown to perform similarly to other methods (e.g., GRU) in many settings [87].”
MethodPan, S. J., & Yang, Q. (2010) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2009.191 · Cited by 23,387 · cited 1×
“The introduction of such methods will reduce the risk of overfitting, but Bayesian models can naturally be adapted to hierarchical models which can naturally lead to transfer learning frameworks [96].”
BackgroundAdvances in neural information processing systems 7
Cios, K., & Shields, M. (1997) · Neurocomputing · doi:10.1016/s0925-2312(97)90018-7 · Cited by 22,059
Multilayer feedforward networks are universal approximators
Hornik, K., Stinchcombe, M., & White, H. (1989) · Neural Networks · doi:10.1016/0893-6080(89)90020-8 · Cited by 21,728 · cited 1×
“However, since these models can model arbitrarily complex functions [46] care must be taken to minimise the risk of overfitting.”
BackgroundQuinlan, J. R. (1986) · Machine Learning · doi:10.1007/bf00116251 · Cited by 21,439 · cited 2×
“We note that whilst research has often focused on which ML algorithm performs best for the given dataset, we will assume instead here that virtually any state-of-the-art ML algorithm (e.g., kernel Support Vector Machines (SVMs) [12], Decision Trees [13], Bayesian classifiers [14]) can be made to perform equivalently g…”
BackgroundConditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
Lafferty, J., McCallum, A., & Pereira, F. C. N. (2001) · ScholarlyCommons (University of Pennsylvania) · Cited by 12,935
Research works printed but not linked: 3
- Shawe-Taylor, J., and Cristianini, N. (2000). Support Vector Machines, Cambridge University Press.
- Goodfellow, I., Bengio, Y., Courville, A., and Bengio, Y. (2016). Deep Learning, MIT Press.
- Tipping, 2001, Sparse Bayesian Learning and the Relevance Vector Machine, J. Mach. Learn. Res., 1, 211
Cited by
Measurement of physical activity in clinical practice using accelerometers
Arvidsson, D., Fridolfsson, J., & Börjesson, M. (2019) · Journal of Internal Medicine · doi:10.1111/joim.12908 · Cited by 239
Self-supervised learning for human activity recognition using 700,000 person-days of wearable data
Yuan, H., Chan, S., Creagh, A. P., Tong, C., Acquah, A., Clifton, D. A., & Doherty, A. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01062-3 · Cited by 166
“Current human activity recognition (HAR) models typically rely on manual feature engineering (Twomey et al., 2018; Haresamudram et al., 2019) partly due to the very limited size of existing labelled datasets.”
A Comprehensive Study of Activity Recognition Using Accelerometers
Twomey, N., Diethe, T., Fafoutis, X., Elsts, A., McConville, R., Flach, P., & Craddock, I. (2018) · doi:10.20944/preprints201803.0147.v1 · Cited by 143
Self-citationHARTH: A Human Activity Recognition Dataset for Machine Learning
Logacjov, A., Bach, K., Kongsvold, A., Bårdstu, H. B., & Mork, P. J. (2021) · Sensors · doi:10.3390/s21237853 · Cited by 184
“Human activity recognition (HAR) is a field of study that focuses on recognizing specific human physical activities and postures based on sensor data [7,8].”
BackgroundAn IoT-Based Design Using Accelerometers in Animal Behavior Recognition Systems
Tran, D.-N., Nguyen, T. N., Khanh, P. C. P., & Tran, D. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3051194 · Cited by 142
“Incredibly we got high per-530 formance when classifying feeding (0.914 accuracy, 0.884 sensitivity, and 0.956 PPV) and standing (0.88 accuracy, 0.928 sensitivity, and 0.842 PPV), which was often a limit [4], [5], [22], [24]–[26], [45]–[47].”
BackgroundInfluentialLARa: Creating a Dataset for Human Activity Recognition in Logistics Using Semantic Attributes
Niemann, F., Reining, C., Rueda, F. M., Nair, N. R., Steffens, J. A., Fink, G. A., & ten Hompel, M. (2020) · Sensors · doi:10.3390/s20154083 · Cited by 120
“[1,11,20,21].”
Human activity recognition with smartphone-integrated sensors: A survey
Dentamaro, V., Gattulli, V., Impedovo, D., & Manca, F. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.123143 · Cited by 118
Human Activity Recognition Based on Dynamic Active Learning
Bi, H., Perello-Nieto, M., Santos-Rodriguez, R., & Flach, P. (2021) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.3013403 · Cited by 81
“According to [33], [62], mid-sized time windows (from 5 to 7 seconds long) perform best from a range of window size from 1 to 15 seconds for wearable sensors.”
BackgroundMethodSelf-citationChambers, R., Yoder, N. C., Carson, A., Junge, C., Allen, D. E., Prescott, L. M., Bradley, S., Wymore, G., Lloyd, K., & Lyle, S. (2021) · Animals · doi:10.3390/ani11061549 · Cited by 84
Baldominos, A., Cervantes, A., Sáez, Y., & Isasi, P. (2019) · Sensors · doi:10.3390/s19030521 · Cited by 86
“It is important to remark that, as found in previous works [38], the experimental setup characterizes the data that is going to be delivered to the classifier, and therefore, some configurations can be more precise than others.”
BackgroundLi, J., Tian, L., Wang, H., An, Y., Wang, K., & Lei, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2905575 · Cited by 89
“This rate is sufficient to capture the motion of the human body because 99% of its energy is contained below 15Hz [40].”
BackgroundBach, K., Kongsvold, A., Bårdstu, H. B., Bardal, E. M., Kjærnli, H. S., Herland, S., Logacjov, A., & Mork, P. J. (2021) · Journal for the Measurement of Physical Behaviour · doi:10.1123/jmpb.2021-0015 · Cited by 85
“Which postures and activity types that can be detected and with which accuracy depends on several factors such as the positioning and number of sensors and the window length used for extracting information from the acceleration signal (Arvidsson et al., 2019; Twomey et al., 2018).”
BackgroundHuman Activity Recognition for Production and Logistics—A Systematic Literature Review
Reining, C., Niemann, F., Rueda, F. M., Fink, G. A., & ten Hompel, M. (2019) · Information · doi:10.3390/info10080245 · Cited by 82
“HAR faces challenges with regards to the settings of the recording environment, number of participants as well as sensors and their configuration [20,29].”
BackgroundHuman activity classification using Decision Tree and Naïve Bayes classifiers
Maswadi, K., Ghani, N. B. A., Hamid, S. B., & Rasheed, M. B. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-020-10447-x · Cited by 88
MultiCNN-FilterLSTM: Resource-efficient sensor-based human activity recognition in IoT applications
Park, H., Kim, N., Lee, G. H., & Choi, J. K. (2022) · Future Generation Computer Systems · doi:10.1016/j.future.2022.09.024 · Cited by 79
A Low Power Fall Sensing Technology Based on FD-CNN
He, J., Zhang, Z., Wang, X., & Yang, S. (2019) · IEEE Sensors Journal · doi:10.1109/jsen.2019.2903482 · Cited by 66
“With the development of microelectromechanical systems (MEMS), researchers have integrated inertial sensors into small wearable devices to achieve fall detection [10], [11].”
BackgroundChan, S., Hang, Y., Tong, C., Acquah, A., Schonfeldt, A., Gershuny, J., & Doherty, A. (2024) · Scientific Data · doi:10.1038/s41597-024-03960-3 · Cited by 61
“For RF and XGBoost, we extracted time-series features from the accelerometry that are commonly used in the literature 30 including time and frequency domain features, angular and peak features, resulting in a total of 40 features per window.”
Gómez-Carmona, O., Casado-Mansilla, D., Kraemer, F. A., López-de-Ipiña, D., & García-Zubia, J. (2020) · Future Generation Computer Systems · doi:10.1016/j.future.2020.06.013 · Cited by 52
“For this reason, we have decided to cut o ff this reduction and keep 16 Hz as the reference downsampled ratio, since this frequency belongs to the 15 Hz–20 Hz range, which is considered as appropriate for HAR systems [61].”
On the Challenges and Potential of Using Barometric Sensors to Track Human Activity
Manivannan, A., Chin, W. C. B., Barrat, A., & Bouffanais, R. (2020) · Sensors · doi:10.3390/s20236786 · Cited by 63
“For instance, three-axis accelerometers are the most widely used sensors [10,11], followed by three-axis gyroscopes and magnetometers.”
MethodA Microservices e-Health System for Ecological Frailty Assessment Using Wearables
García-Moreno, F. M., Bermúdez-Edo, M., Garrido, J. L. G., Rodríguez-García, E., Pérez‐Mármol, J. M., & Rodríguez‐Fórtiz, M. J. (2020) · Sensors · doi:10.3390/s20123427 · Cited by 50
“We tested these algorithms, with R programming language, tuning its specific hyper-parameters based on related studies [45,49,80]:”
MethodSchmeling, L., Elmamooz, G., Hoang, P. T., Kozar, A., Nicklas, D., Sünkel, M., Thurner, S., & Rauch, E. (2021) · Animals · doi:10.3390/ani11092660 · Cited by 49
Chen, P.-W., Baune, N. A., Zwir, I., Wang, J., Swamidass, V., & Wong, A. W. K. (2021) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph18041634 · Cited by 41
Scheduling Computational and Energy Harvesting Tasks in Deadline-Aware Intermittent Systems
Islam, B., & Nirjon, S. (2020) · IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS) · doi:10.1109/rtas48715.2020.00-14 · Cited by 58
An Industrial IoT Solution for Evaluating Workers' Performance Via Activity Recognition
Forkan, A. R. M., Montori, F., Georgakopoulos, D., Jayaraman, P. P., Yavari, A., & Morshed, A. (2019) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs.2019.00139 · Cited by 50
“that a total of 69 activities are recognised in various activity in daily living studies using different IoT sensors or mobile phones containing accelerometer, positioned in different parts of the human body [34].”
BackgroundDeveloping a Smartwatch-Based Healthcare Application: Notes to Consider
Ramezani, R., Cao, M., Earthperson, A., & Naeim, A. (2023) · Sensors · doi:10.3390/s23156652 · Cited by 46
“A sampling rate of 20 Hz has been validated as effective for facilitating the day-long operation of these sensors, ensuring precise activity recognition [32].”
BackgroundDeep Learning for Accelerometric Data Assessment and Ataxic Gait Monitoring
Procházka, A., Dostál, O., Cejnar, P., Mohamed, H. I., Pavelek, Z., Vališ, M., & Vyšata, O. (2021) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2021.3051093 · Cited by 40
PlÖtz, T. (2022) · ACM Computing Surveys · doi:10.1145/3459666 · Cited by 34
“Take, for example, the case of accelerometry-based activity recognition that aims at automatically logging activities of daily living [68].”
BackgroundOnline active learning for human activity recognition from sensory data streams
Mohamad, S., Sayed‐Mouchaweh, M., & Bouchachia, A. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.08.092 · Cited by 41
de Araújo, A. C. A., Santos, E. G. R., de Sá, K. S. G., Furtado, V. K. T., Santos, F. A. F., de Lima, R. C., Krejčová, L. V., Santos‐Lobato, B. L., Pinto, G., Cabral, A. D. S., et al. (2020) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2020.00778 · Cited by 39
“The more common features extracted from inertial readings express amplitude of oscillatory series, their spectral content, regularity, and coherence (Meigal et al., 2012; Twomey et al., 2018).”
BackgroundVita, A. D., Russo, A., Pau, D. P., Benedetto, L. D., Rubino, A., & Licciardo, G. D. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2020.3011984 · Cited by 33
“1) Filtering: Gravity acceleration is estimated from the low-frequency component of the acquired accelerations [45].”
BackgroundHuman Behavior Recognition Model Based on Feature and Classifier Selection
Gao, G., Li, Z., Huan, Z., Chen, Y., Liang, J., Zhou, B., & Dong, C. (2021) · Sensors · doi:10.3390/s21237791 · Cited by 38
“Because 99% of its energy is contained below 15 Hz [45], this rate is sufficient to capture the movement of the human body.”
BackgroundShiwani, M. A., Chico, T., Ciravegna, F., & Mihaylova, L. S. (2023) · Sensors · doi:10.3390/s23125752 · Cited by 40
“Studies using only magnetometers for activity recognition are limited due to their poor performance and power consumption compared to accelerometers and gyroscopes [36].”
Logacjov, A., & Bach, K. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107478 · Cited by 41
A benchmark for domain adaptation and generalization in smartphone-based human activity recognition
Napoli, O. O., Duarte, D., Alves, P., Soto, D. H. P., de Oliveira, H. E., Rocha, A. D. R., Boccato, L., & Borin, E. (2024) · Scientific Data · doi:10.1038/s41597-024-03951-4 · Cited by 52
“Although there are several works that used the fixed duration windowing scheme with values as small as 0.5 seconds to values larger than 30 seconds 2,3,24,27 , the most common range for HAR tasks is between 1–5 seconds, as described in detail in the systematic review work by Straczkiewicz et al .”
BackgroundInfluentialCross-Attention Enhanced Pyramid Multi-Scale Networks for Sensor-Based Human Activity Recognition
Pang, H., Li, Z., & Fang, H. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3377353 · Cited by 34
“As opposed to this, sensor-based HAR systems utilize various motion sensors (e.g., accelerometers and gyroscopes) embedded in wearable or portable devices to collect human motion data [11], [12].”
BackgroundA Framework for Maternal Physical Activities and Health Monitoring Using Wearable Sensors
Ullah, F., Iqbal, A., Iqbal, S., Kwak, D., Anwar, H., Khan, A., Ullah, R., Siddique, H., & Kwak, K. S. (2021) · Sensors · doi:10.3390/s21154949 · Cited by 32
“[58] PA1, PA6, PA7, PA10, PA11 RF: 80%, MLP: 81%”
Significant Features for Human Activity Recognition Using Tri-Axial Accelerometers
Bennasar, M., Price, B. A., Gooch, D., Bandara, A. K., & Nuseibeh, B. (2022) · Sensors · doi:10.3390/s22197482 · Cited by 38
“Furthermore, it has been reported that placing an accelerometer on the wrist is the optimal placement position because it is most preferred by participants [14].”
BackgroundResultInfluentialReal Time Sign Language Translation Systems: A review study
Papatsimouli, M., Kollias, K.-F., Lazaridis, L., Maraslidis, G. S., Michailidis, H., Sarigiannidis, P. G., & Fragulis, G. F. (2022) · International Conference on Modern Circuits and Systems Technologies · doi:10.1109/mocast54814.2022.9837666 · Cited by 32
Classifying sitting, standing, and walking using plantar force data
Merry, K. J., Macdonald, E., MacPherson, M., Aziz, O., Park, E., Ryan, M., & Sparrey, C. J. (2021) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-020-02297-4 · Cited by 31
Are Microcontrollers Ready for Deep Learning-Based Human Activity Recognition?
Elsts, A., & McConville, R. (2021) · Electronics · doi:10.3390/electronics10212640 · Cited by 33
“classification problem from acceleration data [7,8].”
BackgroundMethodInfluentialSelf-citationGarcía-Moreno, F. M., Bermúdez-Edo, M., Rodríguez-García, E., Pérez‐Mármol, J. M., Garrido, J. L. G., & Rodríguez‐Fórtiz, M. J. (2021) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2021.104625 · Cited by 33
ADAM-sense: Anxiety-displaying activities recognition by motion sensors
Khan, N. S., Ghani, M. S., & Anjum, G. (2021) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2021.101485 · Cited by 25
A Personalized Classifier for Human Motion Activities With Semi-Supervised Learning
Singh, G., Chowdhary, M., Kumar, A., & Bahl, R. (2020) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2020.3036277 · Cited by 25
Energy-efficient activity recognition framework using wearable accelerometers
Elsts, A., Twomey, N., McConville, R., & Craddock, I. J. (2020) · Journal of Network and Computer Applications · doi:10.1016/j.jnca.2020.102770 · Cited by 26
Self-citationHeidarivincheh, F., McConville, R., Morgan, C., McNaney, R., Masullo, A., Mirmehdi, M., Whone, A. L., & Craddock, I. J. (2021) · Sensors · doi:10.3390/s21124133 · Cited by 32
“For Acl-RF and Acl-LSTM models, we extract features from the accelerometer data which are frequently used in accelerometer signal processing [52,53].”
MethodSelf-citationHuman Activities Recognition Using Accelerometer and Gyroscope
Ferrari, A., Micucci, D., Mobilio, M., & Napoletano, P. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-34255-5_28 · Cited by 28
SelfPAB: large-scale pre-training on accelerometer data for human activity recognition
Logacjov, A., Herland, S., Ustad, A., & Bach, K. (2024) · Applied Intelligence · doi:10.1007/s10489-024-05322-3 · Cited by 35
“Objective measurement-based human activity recognition (HAR) is a research field focusing on predicting human pos-B tures and physical activities from sensor data [43].”
Chaurasia, S. K., & Reddy, S. R. N. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11410-0 · Cited by 22
“Only accelerometer based survey is discussed in [119] and sensor fusion using multiple sensors is covered in [85].”
MethodThe new design of cows' behavior classifier based on acceleration data and proposed feature set
Khanh, P. C. P., Tran, D., Duong, V. T., Thinh, N. H., & Tran, D.-N. (2020) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2020151 · Cited by 25
“…are realized; if the window length becomes too large, the probability that a given window contains more than one activity is increased, the delay before a classification output can be generated is increased, and the number of training examples for the classification will also be reduced [33].”
BackgroundVesta: A digital health analytics platform for a smart home in a box
McConville, R., Archer, G. T., Craddock, I. J., Kozłowski, M., Piechocki, R. J., Pope, J., & Santos‐Rodríguez, R. (2020) · Future Generation Computer Systems · doi:10.1016/j.future.2020.07.046 · Cited by 23
“For example, Twomey et al. [34] illustrate the vast array of features and algorithms typically used in the task of activity recognition.”
BackgroundSelf-citationClassifying diverse manual material handling tasks using a single wearable sensor
Porta, M., Kim, S., Pau, M., & Nussbaum, M. A. (2021) · Applied Ergonomics · doi:10.1016/j.apergo.2021.103386 · Cited by 28
Shi, C., Wang, Y., Chen, Y. J., Saxena, N., & Wang, C. (2020) · Annual Computer Security Applications Conference · doi:10.1145/3427228.3427259 · Cited by 33
“Previouswork [27, 43] found that human motion-related accelerations usually have frequencies lower than 20Hz.”
BackgroundDynamic Segmentation for Physical Activity Recognition Using a Single Wearable Sensor
Alhammad, N., & Al-Dossari, H. Z. (2021) · Applied Sciences · doi:10.3390/app11062633 · Cited by 28
Tjøsvoll, S. O., Wiggen, Ø. N., Gonzalez-Sanchez, V. E., Seeberg, T. M., Redzovic, S. E., Liaset, I. F., Holtermann, A., & Fimland, M. S. (2022) · Annals of Work Exposures and Health · doi:10.1093/annweh/wxac052 · Cited by 24
“Development in microelectromechanical systems technology enables feasible long-term measurements of physical exposures, with a high level of accuracy (Duncan et al., 2018; Stewart et al., 2018; Twomey et al., 2018).”
BackgroundDriver State and Behavior Detection Through Smart Wearables
Tavakoli, A., Kumar, S., Boukhechba, M., & Heydarian, A. (2021) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv48863.2021.9575431 · Cited by 26
CNN‐LSTM‐Based Late Sensor Fusion for Human Activity Recognition in Big Data Networks
Baloch, Z., Shaikh, F. K., & Unar, M. A. (2022) · Wireless Communications and Mobile Computing · doi:10.1155/2022/3434100 · Cited by 24
The Role of Multi-Sensor Measurement in the Assessment of Movement Quality: A Systematic Review
Swain, T. A., McNarry, M. A., Runacres, A., & Mackintosh, K. A. (2023) · Sports Medicine · doi:10.1007/s40279-023-01905-1 · Cited by 27
Talk, Text, Tag? Understanding Self-Annotation of Smart Home Data from a User’s Perspective
Tonkin, E. L., Burrows, A., Woznowski, P., Laskowski, P., Yordanova, K., Twomey, N., & Craddock, I. J. (2018) · Sensors · doi:10.3390/s18072365 · Cited by 24
“Recent applied research in activity recognition has highlighted a key challenge in the field of autonomous activity recognition: resolving the statistical differences in data collected in the lab and data collected in the wild [19].”
BackgroundSelf-citationFeL-MAR: Federated learning based multi resident activity recognition in IoT enabled smart homes
Dahal, A., Moulik, S., & Mukherjee, R. (2024) · Future Generation Computer Systems · doi:10.1016/j.future.2024.107552 · Cited by 18
Human Action Recognition Using Smartphone Sensors
Saha, A., Sharma, T., Batra, H., Jain, A., & Pal, V. (2020) · International Conference on Computational Performance Evaluation (ComPE) · doi:10.1109/compe49325.2020.9200169 · Cited by 19
“Twomey, Diethe, Fafoutis, et. al [9] researched on a comprehensive s-tudy on activity recognition system that particularly focused on long-term activity recognition in real-world settings.”
BackgroundDerakhshani, S. M., Overduin, M., van Niekerk, T. G. C. M., & Koerkamp, P. W. G. G. (2022) · Animals · doi:10.3390/ani12050536 · Cited by 26
Self-Supervised WiFi-Based Activity Recognition
Bocus, M. J., Lau, H.-S., McConville, R., Piechocki, R. J., & Santos‐Rodríguez, R. (2022) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps56602.2022.10008537 · Cited by 29
“when compared to current technologies such as cameras or wearable sensors [2].”
BackgroundSelf-citationA Survey on Human Activity Recognition Using Accelerometer Sensor
Alrazzak, U., & Alhalabi, B. (2019) · International Conference on Informatics, Electronics & Vision (ICIEV) · doi:10.1109/iciev.2019.8858578 · Cited by 15
Bernal, V. C. E., Hiremath, S. V., Wolf, B., Riley, B., Mendonca, R. J., & Johnson, M. J. (2021) · Journal of Rehabilitation and Assistive Technologies Engineering · doi:10.1177/20556683211044640 · Cited by 17
GestureKeeper: Gesture Recognition for Controlling Devices in IoT Environments
Sideridis, V., Zacharakis, A., Tzagkarakis, G., & Papadopouli, M. (2019) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2019.8903044 · Cited by 21
IoT-Based Patient Movement Monitoring: The Post-Operative Hip Fracture Rehabilitation Model
Gupta, A., & Al‐Anbuky, A. (2021) · Future Internet · doi:10.3390/fi13080195 · Cited by 19
“A recent article surveyed the significance of healthcare IoT from clinical perspectives by discussing its current trends, application demands, and challenges [30].”
BackgroundRecognizing Physical Activities for Spinal Cord Injury Rehabilitation Using Wearable Sensors
Alhammad, N., & Al-Dossari, H. Z. (2021) · Sensors · doi:10.3390/s21165479 · Cited by 15
“For upper-limb activitie , sensors are plac d on he wrist and upper arm to obtain the highest performance [25].”
BackgroundNg, J. Y. Y., Zhang, J. H., Hui, S. S., Jiang, G., Yau, F., Cheng, J. S.-C., & Ha, A. S. C. (2024) · PLoS ONE · doi:10.1371/journal.pone.0299295 · Cited by 15
Hedegaard, M., Anvari‐Moghaddam, A., Jensen, B. K., Jensen, C. B., Pedersen, M. K., & Samani, A. (2019) · Medical Engineering & Physics · doi:10.1016/j.medengphy.2019.10.006 · Cited by 18
Zalewski, P., Marchegiani, L., Elsts, A., Piechocki, R. J., Craddock, I. J., & Fafoutis, X. (2020) · Sensors · doi:10.3390/s20061655 · Cited by 17
“This is a typical observation in activity-based wearable sensors [32].”
BackgroundMethodSelf-citationDeveloping remote patient monitoring infrastructure using commercially available cloud platforms
Cao, M., Ramezani, R., Katakwar, V. K., Zhang, W., Boda, D., Wani, M., & Naeim, A. (2024) · Frontiers in Digital Health · doi:10.3389/fdgth.2024.1399461 · Cited by 20
Dinarević, E. C., Husić, J. B., & Baraković, S. (2019) · Sensors · doi:10.3390/s19235206 · Cited by 12
Discrimination of cycling patterns using accelerometric data and deep learning techniques
Procházka, A., Charvátová, H., Vyšata, O., Jarchi, D., & Sanei, S. (2020) · Neural Computing and Applications · doi:10.1007/s00521-020-05504-3 · Cited by 14
“Computer-assisted monitoring of motion activities [31, 33, 34, 36, 57] allows us to improve the quality of life in many different areas, including health care and personal fitness.”
BackgroundDahal, A., & Moulik, S. (2025) · IEEE Sensors Letters · doi:10.1109/lsens.2025.3526446 · Cited by 12
“In contrast, the work described in [13] applied a random forest (RF) technique, which incorporated the bootstrap aggregation (bagging) method alongside random node selection for the creation of decision trees (DT).”
Lee, S., Lee, D.-W., & Kim, M. S. (2023) · Sensors · doi:10.3390/s23042278 · Cited by 13
“There are various human activity datasets comprising acceleration data, such as WISDM, UCI HAR, and MHEALTH [24], but most of them focus on the change in the human state, not on the transition activity.”
MethodAzadi, B., Haslgrübler, M., Sopidis, G., Murauer, M., Anzengruber, B., & Ferscha, A. (2019) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3316782.3322749 · Cited by 11
“We used inertial sensors, as it has been proposed in different studies [28, 34] that proved that accelerometer signals are helpful in human activity recognition.”
MethodA hierarchical machine learning framework for the identification of automated construction
Harichandran, A., Raphael, B., & Mukherjee, A. (2021) · Journal of Information Technology in Construction · doi:10.36680/j.itcon.2021.031 · Cited by 19
“Some examples can be found in (Catal et al., 2015; Akhavian and Behzadan, 2016; Cheng et al., 2017; Twomey et al., 2018).”
Holmes, M., Perelló-Nieto, M., Song, H., Tonkin, E. L., Grant, S., & Flach, P. A. (2020) · Journal of Healthcare Informatics Research · doi:10.1007/s41666-020-00072-6 · Cited by 14
Self-citationOAFPM: optimized ANFIS using frequent pattern mining for activity recognition
Marimuthu, P., Perumal, V., & Vijayakumar, V. (2019) · The Journal of Supercomputing · doi:10.1007/s11227-019-02802-z · Cited by 11
“The real-world problems in implementing and evaluation of various activity recognition (AR) models using accelerometer have been discussed in [23].”
BackgroundPoli, A., Spinsante, S., Nugent, C., & Cleland, I. (2019) · Computer communications and networks · doi:10.1007/978-3-030-25590-9_7 · Cited by 10
Rethinking IoT Network Reliability in the Era of Machine Learning
Fafoutis, X., & Marchegiani, L. (2019) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings/greencom/cpscom/smartdata.2019.00189 · Cited by 10
“This is a commonly studied problem with numerous applications, such as assisted living for the elderly and long-term behavioural analytics for individuals that suffer from chronic illness [19].”
BackgroundResultSelf-citationPearce, J., Chang, Y., Xia, D., & Abeyesinghe, S. M. (2024) · Animals · doi:10.3390/ani14131957 · Cited by 11
“However, a higher percentage overlap results in an increase in computational burden and appropriate window length, and overlap should therefore be taken into consideration [41].”
BackgroundInfluentialLuzum, G., Gunnes, M., Lydersen, S., Saltvedt, I. T., Tan, X., Thingstad, P., Thrane, G., & Askim, T. (2023) · Physical Therapy · doi:10.1093/ptj/pzad092 · Cited by 9
“40 For comprehensive examination of the patterns of time spent being physically active, the cumulative bouts spent in upright behaviors or walking were assigned to 6 predefined zones: zone 1 (3 seconds to 5 minutes) and zone 2 (5 minutes to 10 minutes) were defined as short-bout PA, while zone…”
Deep Architectures for Human Activity Recognition using Sensors
Baloch, Z., Shaikh, F. K., & Unar, M. A. (2019) · 3C Tecnología_Glosas de innovación aplicadas a la pyme · doi:10.17993/3ctecno.2019.specialissue2.14-35 · Cited by 10
Yamagata, T., Santos‐Rodríguez, R., McConville, R., & Elsts, A. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.06134 · Cited by 9
“There has been much work in the general area of activity recognition [7], particularly wrist-based accelerometers, due to their performance and acceptability to users [8].”
BackgroundSelf-citationThe Multimodel Stacking and Ensemble Framework for Human Activity Recognition
Dahal, A., & Moulik, S. (2024) · IEEE Sensors Letters · doi:10.1109/lsens.2024.3451960 · Cited by 8
Latent Space Representation of Human Movement: Assessing the Effects of Fatigue
Rousseau, T., Venture, G., & Hernandez, V. (2024) · Sensors · doi:10.3390/s24237775 · Cited by 8
“Human Activity Recognition (HAR) is a specialized research area that combines machine learning and sensor data to analyze and monitor human movements [25,26].”
Systematic design of health monitoring systems centered on older adults and ADLs
García-Moreno, F. M., Bermúdez-Edo, M., Pérez‐Mármol, J. M., Garrido, J. L. G., & Rodríguez‐Fórtiz, M. J. (2024) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-024-02432-3 · Cited by 9
“To classify the frailty of the elders, we applied four machine learning techniques with the hyper-parameters tuned as in the literature [50–52] shown in Fig.”
MethodDetecting Signatures of Early-stage Dementia with Behavioural Models Derived from Sensor Data
Poyiadzi, R., Yang, W., Ben‐Shlomo, Y., Craddock, I. J., Coulthard, E. J., Santos‐Rodríguez, R., Selwood, J., & Twomey, N. (2020) · Explore Bristol Research · doi:10.48550/arxiv.2007.03615 · Cited by 11
“50 seconds following [33]).”
Self-citationMachine Learning for Sensing Applications: A Tutorial
Shirmohammadli, V., & Bahreyni, B. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3112901 · Cited by 8
Emotions and Activity Recognition System Using Wearable Device Sensors
Rumiantcev, M. (2021) · Conference of Open Innovations Association (FRUCT) · doi:10.23919/fruct50888.2021.9347652 · Cited by 8
Manivannan, A., Willemse, E. J., Balamurali, B. T., Chin, W. C. B., Zhou, Y., Tunçer, B., Barrat, A., & Bouffanais, R. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3157806 · Cited by 8
From Human Pose to On-Body Devices for Human-Activity Recognition
Rueda, F. M., & Fink, G. A. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412283 · Cited by 11
“These features are, for example, the mean, variance, and correlation for the time domain, or the entropy, energy, and coherence extracted from the Fourier Transform in the frequency domain [1], [10], [13], [23].”
BackgroundMethodInfluentialGuo, P., & Nakayama, M. (2025) · Sensors · doi:10.3390/s25226988 · Cited by 6
“For human-scale activities (e.g., walking, running, jumping), most discriminative information lies in low-frequency bands, typically well below 10–15 Hz [42].”
Ward, S., Hu, S., & Zecca, M. (2023) · Sensors · doi:10.3390/s23031416 · Cited by 6
“Accelerometers have been demonstrated to be effective at identifying a wide range of human activities [10].”
BackgroundRoy, A., Dutta, H., Bhuyan, A. K., & Biswas, S. K. (2024) · Sensors · doi:10.3390/s24144444 · Cited by 6
Interactive Game-Based Motor Rehabilitation Using Hybrid Sensor Architecture
Ghosh, A., & Saha, S. (2020) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-5225-9643-1.ch015 · Cited by 6
Development of Automated Top-Down Construction System for Low-rise Building Structures
Harichandran, A., Raphael, B., & Mukherjee, A. (2020) · International Journal of Industrialized Construction · doi:10.29173/ijic217 · Cited by 5
Wireless Sensor and Actuator Networks for Smart Cities
Kantarci, B., & Oktuğ, S. (2018) · MDPI eBooks · doi:10.3390/books978-3-03897-424-6 · Cited by 6
Keyvanpour, M. R., Mehrmolaei, S., Shojaeddini, S. V., & Esmaeili, F. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-16795-8 · Cited by 4
“Yet independence in placement and orientation [54], convenient use, accurate activity prediction [133], and smaller size and low cost [170] are regarded as positive points.”
Metadata
- Authors
- Twomey, N., Diethe, T., Fafoutis, X., Elsts, A., McConville, R., Flach, P., & Craddock, I.
- Affiliations
- University of Bristol (GB)
- Corresponding author
- Twomey, Niall
- Diethe, Tom
- Venue
- Informatics
- Publisher
- MDPI AG
- Published
- 2018-05-30
- Volume
- 5
- Issue
- 2
- Pages
- 27
- ISSN
- 2227-9709 (ISSN-L)
- OpenAlex
- W2790414609
- Cited by
- 143
- References
- 95
- Citations by year
- 2026: 1 · 2025: 9 · 2024: 20 · 2023: 10 · 2022: 19 · 2021: 30 · 2020: 24 · 2019: 26 · 2018: 4
- FWCI
- 10.50
- Type
- journal article
- Language
- en
- Open access
- gold
- dx.doi.org
- Licence
- Locations
- Informatics (publishedVersion, cc-by)
- Preprints.org (acceptedVersion, cc-by)
- Explore Bristol Research (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Bristol Research (University of Bristol) (submittedVersion, cc-by)
- Funders
- Horizon 2020 Framework Programme
- Medical Research Council (MC_PC_16029)
- Engineering and Physical Sciences Research Council (EP/K031910/1, H2020)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Physical Activity and Health
- Social Sciences › Transportation › Human Mobility and Location-Based Analysis
- Keywords
- activity recognition, accelerometer, wearable sensors, accelerometer placement, time series segmentation, feature extraction, activity selection, test-retest reliability, generalization, field deployment
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.
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.