Applying Machine Learning for Sensor Data Analysis in Interactive Systems: Common Pitfalls of Pragmatic Use and Ways to Avoid Them
The DOI matches its Crossref record; 54 of the 55 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
With the widespread proliferation of (miniaturized) sensing facilities and the massive growth and popularity of the field of machine learning (ML) research, new frontiers in automated sensor data analysis have been explored that lead to paradigm shifts in many application domains. In fact, many practitioners now employ and rely more and more on ML methods as integral part of their sensor data analysis workflows—thereby not necessarily being ML experts or having an interest in becoming one. The availability of toolkits that can readily be used by practitioners has led to immense popularity and widespread adoption and, in essence, pragmatic use of ML methods. ML having become mainstream helps pushing the core agenda of practitioners, yet it comes with the danger of misusing methods and as such running the risk of leading to misguiding if not flawed results. Based on years of observations in the ubiquitous and interactive computing domain that extensively relies on sensors and automated sensor data analysis, and on having taught and worked with numerous students in the field, in this article I advocate a considerate use of ML methods by practitioners, i.e., non-ML experts, and elaborate on pitfalls of an overly pragmatic use of ML techniques. The article not only identifies and illustrates the most common issues, it also offers ways and practical guidelines to avoid these, which shall help practitioners to benefit from employing ML in their core research domains and applications.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“The primary metric of evaluation is the mean (or unweighted) F1-score, as it is more resistant to class imbalance, which commonly occurs in wearable datasets [151].”
From Past, Present, and Future of Sensor-based Human Activity Recognition Using Wearables: A Surveying Tutorial on a Still Challenging Task (2025) · cites this paper as [151]
In the citing paper
“The accuracy and weighted F1-score are not utilized as they are affected by the skewed class distributions, whereas the unweighted mean F1-score, while not ideal, is a reasonable strategy [53].”
From Assessing the State of Self-Supervised Human Activity Recognition Using Wearables (2022) · cites this paper as [53]
In the citing paper
“We report the mean of the test set macro F1-scores across 5 folds and 5 random runs (to account for class-imbalance [26]).”
From How Much Unlabeled Data is Really Needed for Effective Self-Supervised Human Activity Recognition? (2023) · cites this paper as [26]
Citations
Cites
Interrater reliability: The kappa statistic
McHugh, M. L. (2012) · Biochemia Medica · doi:10.11613/bm.2012.031 · Cited by 19,877 · cited 1×
“Annotator agreement can be quantized through specific measures, such as kappa statistics (interrater reliability) [50].”
BackgroundRepresentation Learning: A Review and New Perspectives
Bengio, Y., Courville, A., & Vincent, P. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.50 · Cited by 13,001 · cited 1×
“ML methods learn representations [5] and models from example data, thereby aiming for generalization beyond those training data such that the resulting models and methods can effectively be used for accurate analysis of unseen, yet related data.”
MethodScikit-learn: Machine Learning in Python
(2011) · Journal of Machine Learning Research · doi:10.5555/1953048.2078195 · Cited by 8,539
Simultaneous Statistical Inference
Miller, R. G. (1981) · Springer Series in Statistics · doi:10.1007/978-1-4613-8122-8 · Cited by 5,604
Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation
Powers, D. M. W. (2020) · arXiv · doi:10.48550/arxiv.2010.16061 · Cited by 5,440
Understanding and Using Context
Dey, A. K. (2001) · Personal and Ubiquitous Computing · doi:10.1007/s007790170019 · Cited by 5,012 · cited 1×
“The main application case for ML in such settings is the analysis of sensor data for capturing and assessing the context a user is in [17].”
BackgroundA survey of cross-validation procedures for model selection
Arlot, S., & Célisse, A. (2010) · LillOA (Université de Lille (University Of Lille)) · Cited by 3,669
A survey of cross-validation procedures for model selection
Arlot, S., & Celisse, A. (2010) · Statistics Surveys · doi:10.1214/09-ss054 · Cited by 3,522 · cited 1×
“In such cases, researchers often resort to an approximative protocol, namely, to k-fold cross validation [3].”
BackgroundA few useful things to know about machine learning
Domingos, P. (2012) · Communications of the ACM · doi:10.1145/2347736.2347755 · Cited by 3,326 · cited 1×
“There have been efforts within the core ML research community to illustrate what ML techniques can do [20], as well as to raise awareness of what they cannot [70].”
BackgroundGeneralizing from a Few Examples: A Survey on Few-shot Learning
Wang, Y., Yao, Q., Kwok, J. T., & Ni, L. M. (2021) · ACM Computing Surveys · doi:10.1145/3386252 · Cited by 3,165
Location systems for ubiquitous computing
Hightower, J., & Borriello, G. (2001) · Computer · doi:10.1109/2.940014 · Cited by 2,987 · cited 1×
“An important context factor is location, and a range of ML techniques have been developed and deployed to infer the location of a user [33], to predict future locations and purpose of trips [23], and to infer the semantic meaning of locations [42].”
An empirical comparison of supervised learning algorithms
Caruana, R., & Niculescu-Mizil, A. (2006) · Proceedings of the 23rd international conference on Machine learning - ICML '06 · doi:10.1145/1143844.1143865 · Cited by 2,780 · cited 1×
“…developed over the years, which can be categorized into two main classes: (i) supervised learning, i.e., methods that derive models from example data and their associated class annotations [11]; and (ii) unsupervised learning, i.e., methods that derive models purely from unlabelled sample data [4].”
BackgroundA Survey of Indoor Localization Systems and Technologies
Zafari, F., Gkelias, A., & Leung, K. K. (2019) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2019.2911558 · Cited by 2,530 · cited 1×
“general characteristics of the sensor data analysis and as such of the ML methods used are similar [75].”
MethodExplaining Explanations: An Overview of Interpretability of Machine Learning
Gilpin, L. H., Bau, D., Yuan, B. Z., Bajwa, A., Specter, M., & Kagal, L. (2018) · 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA) · doi:10.1109/dsaa.2018.00018 · Cited by 2,507 · cited 1×
“” Note, however, that the wider ML research community has now started to move towards studying intelligibility of ML methods [25].”
BackgroundOrange: data mining toolbox in python
Demšar, J., Curk, T., Erjavec, A., Gorup, C., Hočevar, T., Milutinovic, M., Možina, M., Polajnar, M., Toplak, M., Staric, A., et al. (2013) · Journal of machine learning research · doi:10.5555/2567709.2567736 · Cited by 2,246 · cited 2×
“Examples of popular ML toolkits are Matlab [49], Weka [22], scikit-learn [59], Orange [16] and—more recently—the various deep learning frameworks (e.”
BackgroundMachine Learning: An Applied Econometric Approach
Mullainathan, S., & Spiess, J. (2017) · Journal of Economic Perspectives · doi:10.1257/jep.31.2.87 · Cited by 1,989 · cited 1×
“In recent years the field of ML has seen an explosion in growth and sophisticated methods now do exist that are key enablers for many application areas, such as, Natural Language Processing [36], Computer Vision [65, 69], or finance [54], to name but a few.”
BackgroundA tutorial on human activity recognition using body-worn inertial sensors
Bulling, A., Blanke, U., & Schiele, B. (2014) · ACM Computing Surveys · doi:10.1145/2499621 · Cited by 1,668 · cited 12×
“The typical reflex is to employ some form of the standard activity recognition pipeline (summarized, for example, in Reference [10]): Record and annotate example data, extract analysis frames through a sliding window procedure, extract (heuristic) features, and train a classifier.”
BackgroundMethodInfluentialBarlow, H. B. (1989) · Neural Computation · doi:10.1162/neco.1989.1.3.295 · Cited by 1,650 · cited 1×
“…developed over the years, which can be categorized into two main classes: (i) supervised learning, i.e., methods that derive models from example data and their associated class annotations [11]; and (ii) unsupervised learning, i.e., methods that derive models purely from unlabelled sample data [4].”
BackgroundDoherty, A., Jackson, D., Hammerla, N., Plötz, T., Olivier, P., Granat, M. H., White, T., van Hees, V. T., Trenell, M. I., Owen, C. G., et al. (2017) · PLOS ONE · doi:10.1371/journal.pone.0169649 · Cited by 1,250 · cited 1×
“In scenarios where high sampling rates are used and long-term recordings are analyzed (e.g., Reference [18]), reporting results at sample-level may not be appropriate as the typically dominant null class will lead to inflated results that are not very informative.”
BackgroundSelf-citationProceedings of the International Joint Conference on Artificial Intelligence (IJCAI 2017)
Amendola, G., Ricca, F., & Truszczyński, M. (2017) · International Joint Conference on Artificial Intelligence · Cited by 1,179
Sensor-Based Activity Recognition
Chen, L., Hoey, J., Nugent, C. D., Cook, D. J., & Yu, Z. (2012) · IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) · doi:10.1109/tsmcc.2012.2198883 · Cited by 1,074 · cited 1×
“Activities are another important context factor and a wealth of sensing approaches and especially ML methods for sensor-based human activity recognition (HAR) have been developed and deployed [10, 13].”
BackgroundIntroduction to Semi-Supervised Learning
Zhu, X., & Goldberg, A. B. (2009) · Synthesis lectures on artificial intelligence and machine learning · doi:10.2200/s00196ed1v01y200906aim006 · Cited by 1,022
Understanding the Hawthorne effect
Sedgwick, P., & Greenwood, N. (2015) · BMJ · doi:10.1136/bmj.h4672 · Cited by 824 · cited 1×
“…scripted interactions hardly qualify as “natural” and following a protocol certainly has an effect on the participant’s behavior, ranging from observer (Hawthorne [66]) effect to plain mismatches of regular routines that are not captured by even the most sophisticated data collection protocols.”
BackgroundGeneralizing from a Few Examples: A Survey on Few-Shot Learning
Wang, Y., Yao, Q., Kwok, J. T.-Y., & Ni, L. M.-S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.05046 · Cited by 772
The Opportunity challenge: A benchmark database for on-body sensor-based activity recognition
Chavarriaga, R., Sagha, H., Calatroni, A., Digumarti, S. T., Tröster, G., Millán, J. D. R., & Roggen, D. (2013) · Pattern Recognition Letters · doi:10.1016/j.patrec.2012.12.014 · Cited by 742 · cited 1×
“, Reference [53]) it is often very challenging to obtain sample data annotation in such scenarios, and thus it may be tempting to record data in scripted user studies—a practice that is employed rather frequently [12, 64].”
BackgroundThe psychology of science: a reconnaissance
Maslow, A. H. (1966) · CERN Document Server (European Organization for Nuclear Research) · Cited by 727
Biernacki, C., Celeux, G., & Govaert, G. (2003) · Computational Statistics & Data Analysis · doi:10.1016/s0167-9473(02)00163-9 · Cited by 658 · cited 1×
“As an example, it is well-known that the convergence of the Expectation Maximization (EM) algorithm heavily depends on a reasonable initialization [6].”
BackgroundHöst, M., Regnell, B., & Wohlin, C. (2000) · Empirical Software Engineering · doi:10.1023/a:1026586415054 · Cited by 635 · cited 8×
“The typical reflex is to employ some form of the standard activity recognition pipeline (summarized, for example, in Reference [10]): Record and annotate example data, extract analysis frames through a sliding window procedure, extract (heuristic) features, and train a classifier.”
MethodInfluentialThe Elements of Statistical Learning
Sapatinas, T. (2004) · Journal of the Royal Statistical Society Series A: Statistics in Society · doi:10.1111/j.1467-985x.2004.298_11.x · Cited by 635
Time Series: Data Analysis and Theory
Brillinger, D. R. (2001) · Society for Industrial and Applied Mathematics eBooks · doi:10.1137/1.9780898719246 · Cited by 547
Ensembles of Deep LSTM Learners for Activity Recognition using Wearables
Guan, Y., & Plötz, T. (2017) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3090076 · Cited by 511 · cited 1×
“…only small labeled datasets available for model training and validation [44, 71]; where the quality of the data to be analyzed varies including missing, noisy, and ambiguous data [26]; and where hard real-time and resource constraints need to be met for applications in wearable computing scenarios.”
BackgroundSelf-citationThe Role of Occam's Razor in Knowledge Discovery
Domingos, P. (1999) · Data Mining and Knowledge Discovery · doi:10.1023/a:1009868929893 · Cited by 459 · cited 1×
“In essence, this approach applies Occam’s razor, and thus good scientific practice [19, 63].”
MethodDeepX: A Software Accelerator for Low-Power Deep Learning Inference on Mobile Devices
Lane, N. D., Bhattacharya, S., Georgiev, P., Forlivesi, C., Jiao, L., Qendro, L., & Kawsar, F. (2016) · 2016 15th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN) · doi:10.1109/ipsn.2016.7460664 · Cited by 450 · cited 1×
“Sensor data analysis is challenging and as such the requirements and constraints on ML methods are substantial [60]: (i) sensor data are notoriously noisy, complicating model training; (ii) typically only small amounts of labelled sample data exist, which impacts supervised learning methods; (iii) annotation is often…”
BackgroundUbiquitous Computing Fundamentals
(2018) · Chapman & Hall/CRC eBooks · doi:10.1201/9781420093612 · Cited by 253 · cited 1×
“The vast majority of such systems, which include mobile, wearable, ubiquitous (or pervasive), and interactive computing [41], is based and relies on various physical or virtual sensors.”
BackgroundDiscovering routines from large-scale human locations using probabilistic topic models
Farrahi, K., & Gatica-Perez, D. (2011) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/1889681.1889684 · Cited by 244 · cited 1×
“An important context factor is location, and a range of ML techniques have been developed and deployed to infer the location of a user [33], to predict future locations and purpose of trips [23], and to infer the semantic meaning of locations [42].”
Pattern recognition and machine learning
Abramson, N., Braverman, D., & Sebestyen, G. (1963) · IEEE Transactions on Information Theory · doi:10.1109/tit.1963.1057854 · Cited by 196
Performance metrics for activity recognition
Ward, J. A., Lukowicz, P., & Gellersen, H. W. (2011) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/1889681.1889687 · Cited by 182 · cited 1×
“Instead, one could consider reporting results at event level [72] as in many practical applications it is of interest to report how well specific instances of activities can be predicted rather than focusing on arbitrary sample precision.”
BackgroundActivity logging using lightweight classification techniques in mobile devices
Martín, H., Bernardos, A. M., Iglesias, J., & Casar, J. R. (2013) · Personal and Ubiquitous Computing · doi:10.1007/s00779-012-0515-4 · Cited by 180 · cited 1×
“Prominent examples are innovative and more intuitive input modalities such as voice and gesture [37], or automated activity logging and analysis [47].”
BackgroundKwon, H., Tong, C., Haresamudram, H., Gao, Y., Abowd, G. D., Lane, N. D., & Plötz, T. (2020) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3411841 · Cited by 152
Self-citationA Comprehensive Study of Activity Recognition Using Accelerometers
Twomey, N., Diethe, T., Fafoutis, X., Elsts, A., McConville, R., Flach, P., & Craddock, I. (2018) · Informatics · doi:10.3390/informatics5020027 · Cited by 143 · cited 1×
“Take, for example, the case of accelerometry-based activity recognition that aims at automatically logging activities of daily living [68].”
BackgroundMachine Learning: A Guide to Current Research
Mitchell, T. M., Carbonell, J. G., & Michalski, R. S. (1986) · Kluwer international series in engineering and computer science · doi:10.1007/978-1-4613-2279-5 · Cited by 143
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
Hammerla, N. Y., Kirkham, R., Andras, P., & Ploetz, T. (2013) · Proceedings of the 2013 International Symposium on Wearable Computers · doi:10.1145/2493988.2494353 · Cited by 138 · cited 1×
“For example, in many recognition scenarios it is sufficient to use distribution-based sensor data representations like ECDF—that can be computed with linear time complexity [29]—over, for example, feature learning [31, 61].”
BackgroundSelf-citationPlacer: semantic place labels from diary data
Krumm, J. C., & Rouhana, D. (2013) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/2493432.2493504 · Cited by 115 · cited 1×
“An important context factor is location, and a range of ML techniques have been developed and deployed to infer the location of a user [33], to predict future locations and purpose of trips [23], and to infer the semantic meaning of locations [42].”
Deep Learning for Human Activity Recognition in Mobile Computing
Plotz, T., & Guan, Y. (2018) · Computer · doi:10.1109/mc.2018.2381112 · Cited by 99 · cited 2×
“which there are many: noisy sensor readings; often ambiguous, sometimes erroneous annotation (labeling [43]); small sample datasets; hard real-time constraints for analysis; and so on [60].”
BackgroundExamining the effect of initialization strategies on the performance of Gaussian mixture modeling
Shireman, E. M., Steinley, D., & Brusco, M. J. (2015) · Behavior Research Methods · doi:10.3758/s13428-015-0697-6 · Cited by 93 · cited 1×
“Instead, one could start with a more conventional clustering approach, such as k-means, and use the resulting codebook as initialization for the actual EM optimization step(s) [67].”
MethodOn the role of features in human activity recognition
Haresamudram, H., Anderson, D. V., & Plötz, T. (2019) · Proceedings of the 23rd International Symposium on Wearable Computers · doi:10.1145/3341163.3347727 · Cited by 90 · cited 1×
“For example, in many recognition scenarios it is sufficient to use distribution-based sensor data representations like ECDF—that can be computed with linear time complexity [29]—over, for example, feature learning [31, 61].”
BackgroundSelf-citationBootstrapping Personalised Human Activity Recognition Models Using Online Active Learning
Miu, T., Missier, P., & Plotz, T. (2015) · 2015 IEEE International Conference on Computer and Information Technology; Ubiquitous Computing and Communications; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and Computing · doi:10.1109/cit/iucc/dasc/picom.2015.170 · Cited by 54 · cited 9×
“The typical reflex is to employ some form of the standard activity recognition pipeline (summarized, for example, in Reference [10]): Record and annotate example data, extract analysis frames through a sliding window procedure, extract (heuristic) features, and train a classifier.”
BackgroundInfluentialSelf-citationHandling annotation uncertainty in human activity recognition
Kwon, H., Abowd, G. D., & Plötz, T. (2019) · Proceedings of the 23rd International Symposium on Wearable Computers · doi:10.1145/3341163.3347744 · Cited by 52 · cited 2×
“…reliable methods that can cope with the challenges of real-world applications and systems, of 134:2 T. Plötz which there are many: noisy sensor readings; often ambiguous, sometimes erroneous annotation (labeling [43]); small sample datasets; hard real-time constraints for analysis; and so on [60].”
MethodSelf-citationReal-Time Machine Learning: The Missing Pieces
Nishihara, R., Moritz, P., Wang, S., Tumanov, A., Paul, W., Schleier-Smith, J., Liaw, R., Niknami, M., Jordan, M. I., & Stoica, I. (2017) · Proceedings of the 16th Workshop on Hot Topics in Operating Systems · doi:10.1145/3102980.3102998 · Cited by 49 · cited 1×
“Sensor data analysis is challenging and as such the requirements and constraints on ML methods are substantial [60]: (i) sensor data are notoriously noisy, complicating model training; (ii) typically only small amounts of labelled sample data exist, which impacts supervised learning methods; (iii) annotation is often…”
BackgroundSome of my Best Friends are Linguists
Jelinek, F. (2005) · Computers and the Humanities · doi:10.1007/s10579-005-2693-4 · Cited by 44 · cited 1×
“Mercer (IBM Research) in the context of Hidden Markov Models for automated speech recognition and the relationship between their classification accuracy and the size of the training dataset [39].”
BackgroundAutomatic correction of annotation boundaries in activity datasets by class separation maximization
Kirkham, R., Khan, A., Bhattacharya, S., Hammerla, N. Y., Mellor, S., Roggen, D., & Ploetz, T. (2013) · Ubiquitous Computing · doi:10.1145/2494091.2495988 · Cited by 18 · cited 1×
“For supervised learning approaches, such as the widely used, standard activity recognition chain [10], one would typically need sample-precise annotation or otherwise runs the risk of hurting both model training and evaluation by compromised sample data [40].”
MethodUbiquitous Computing Field Studies
Brush, A. J. B. (2009) · Chapman & Hall/CRC eBooks · doi:10.1201/9781420093612.ch4 · Cited by 16
少数の例からの一般化 少数ショット学習に関する調査【JST・京大機械翻訳】
Yaqing, W., Yao, Q., James, T., & Lionel, M. N. (2020) · ACM Computing Surveys · Cited by 9
Research works printed but not linked: 22
- Abowd Gregory D., 2012, Proceedings of the International Conference on Ubiquitous Computing (UbiComp’12)
- Cohn Gabe, InProceedings of the International Conference on Pervasive Computing (Pervasive’10)
- Eibe Frank, 2016, Witten
- Froehlich Jon, InProceedings of the International Conference on Ubiquitous Computing (UbiComp’09)
- Hammerla Nils, 2015, Proceedings of the AAAI Conference on Artificial Intelligence (AAAI’15)
- Hill T., 2007, STATISTICS: Methods and Applications. StatSoft
- Indurkhya Nitin, 2010, Damerau
- Jacko Julie A., Evolving Technologies, and Emerging Applications
- Jain A. K., Classification Pattern Recognition and Reduction of Dimensionality. Handbook of Statistics, 2
- Li H., Proceedings of the International Symposium on Wearable Computing (ISWC’18)
- MATLAB., version 7.10.0 (R2010a)
- Duda Richard O., 2001, Stork
- Paszke Adam, Advances in Neural Information Processing Systems 32, H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc
- Patel Shwetak N., InProceedings of the International Conference on Ubiquitous Computing (UbiComp’07)
- Ploetz Thomas, 2011, Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI’11)
- Rasmussen Carl Edward, 2001, Proceedings of the International Conference on Advances in Neural Information Processing Systems. 294--300
- Reiss A., Proceedings of the International Symposium on Wearable Computing (ISWC’12)
- Sebe Nicu, 2005, Huang
- Voulodimos Athanasios, 2018, Deep learning for computer vision: A brief review. Comput. Intell. Neurosci
- Wagstaff Kiri L., 2012, Proceedings of the International Conference on Machine Learning (ICML’12)
- Weiser M., 1991, The computer for the 21st century. Sci. Amer. 265, 3
- Witten Ian H., 2016, Pal
Cited by
Don’t push the button! Exploring data leakage risks in machine learning and transfer learning
Apicella, A., Isgrò, F., & Prevete, R. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11326-3 · Cited by 217
ColloSSL: Collaborative Self-Supervised Learning for Human Activity Recognition
Jain, Y., Tang, C. I., Min, C., Kawsar, F., & Mathur, A. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3517246 · Cited by 113
Assessing the State of Self-Supervised Human Activity Recognition Using Wearables
Haresamudram, H., Essa, I., & Plötz, T. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3550299 · Cited by 107
Self-citationFLAME: Federated Learning across Multi-device Environments
Cho, H., Mathur, A., & Kawsar, F. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3550289 · Cited by 79
Cross-Domain HAR: Few-Shot Transfer Learning for Human Activity Recognition
Thukral, M., Haresamudram, H., & Plötz, T. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3704921 · Cited by 70
Self-citation
Metadata
- Authors
- PlÖtz, T.
- Affiliations
- Georgia Institute of Technology (US)
- Corresponding author
- PlÖtz, Thomas
- Venue
- ACM Computing Surveys
- Published
- 2021-07-13
- Volume
- 54
- Issue
- 6
- Pages
- 1-25
- ISSN
- 0360-0300 (ISSN-L)
- 1557-7341
- OpenAlex
- W3180073572
- Cited by
- 34
- References
- 53
- Citations by year
- 2026: 3 · 2025: 4 · 2024: 10 · 2023: 6 · 2022: 9 · 2021: 2
- FWCI
- 2.43
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- ACM Computing Surveys (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Data Visualization and Analytics
- Physical Sciences › Computer Science › Human-Computer Interaction › Interactive and Immersive Displays
- Keywords
- machine learning, sensor data analysis, interactive systems, ubiquitous computing
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.