{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,13]],"date-time":"2026-02-13T15:00:29Z","timestamp":1770994829759,"version":"3.50.1"},"reference-count":29,"publisher":"Walter de Gruyter GmbH","issue":"1","license":[{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,6,24]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>In the last decade, many researchers applied shallow and deep networks for human activity recognition (HAR). Currently, the trending research line in HAR is applying deep learning to extract features and classify activities from raw data. However, we observed that, authors of previous studies have not performed an efficient hyperparameter search on their artificial neural network (shallow or deep)-based classifier. Therefore, in this article, we demonstrate the effect of the random and Bayesian parameter search on a shallow neural network using five HAR databases. The result of this work shows that a shallow neural network with correct parameter optimization can achieve similar or even better recognition accuracy than the previous best deep classifier(s) on all databases. In addition, we draw conclusions about the advantages and disadvantages of the two hyperparameter search techniques according to the results.<\/jats:p>","DOI":"10.1515\/comp-2020-0227","type":"journal-article","created":{"date-parts":[[2021,6,24]],"date-time":"2021-06-24T20:25:49Z","timestamp":1624566349000},"page":"411-422","source":"Crossref","is-referenced-by-count":7,"title":["The effect of hyperparameter search on artificial neural network in human activity recognition"],"prefix":"10.1515","volume":"11","author":[{"given":"Jozsef","family":"Suto","sequence":"first","affiliation":[{"name":"Department of IT Systems and Networks, Faculty of Informatics, University of Debrecen , Debrecen , 4028 , Hungary"}]}],"member":"374","published-online":{"date-parts":[[2021,6,24]]},"reference":[{"key":"2022020121510279073_j_comp-2020-0227_ref_001","doi-asserted-by":"crossref","unstructured":"O. D. Lara, M. A. Labrador, \u201cA survey on human activity recognition using wearable sensors,\u201d IEEE Commun. Surv. Tut., vol. 15, pp. 1192\u20131209, 2013","DOI":"10.1109\/SURV.2012.110112.00192"},{"key":"2022020121510279073_j_comp-2020-0227_ref_002","doi-asserted-by":"crossref","unstructured":"Y. Saez, A. Baldominos, P. Isasi, \u201cA comparison study of classifier algorithms for cross-person physical activity recognition,\u201d Sensors, vol. 17, p. 66, 2017.","DOI":"10.3390\/s17010066"},{"key":"2022020121510279073_j_comp-2020-0227_ref_003","unstructured":"P. Y. Simard, D. Steinkraus, J. C. Platt, \u201cBest practice for convolutional neural networks applied to visual document analysis,\u201d In: 7th International Conference on Document Analysis and Recognition (6 Aug. 2003, Washington, USA), Washington, 2003, pp. 958\u2013962."},{"key":"2022020121510279073_j_comp-2020-0227_ref_004","doi-asserted-by":"crossref","unstructured":"F. J. Ordonez, D. Roggen, \u201cDeep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition,\u201d Sensors, vol. 16, pp. 1\u201325, 2016.","DOI":"10.3390\/s16010115"},{"key":"2022020121510279073_j_comp-2020-0227_ref_005","doi-asserted-by":"crossref","unstructured":"M. Zeng, T. Yu, X. Wang, T. L. Nguyen, O. Mengshoel, \u201cSemi-supervised convolutional neural networks for human activity recognition,\u201d In: 2017 IEEE International Conference on Big Data (11\u201314 Dec. 2017, Boston, USA), Boston, 2017, pp. 522\u2013529.","DOI":"10.1109\/BigData.2017.8257967"},{"key":"2022020121510279073_j_comp-2020-0227_ref_006","doi-asserted-by":"crossref","unstructured":"W. Jiang, Z. Yin, \u201cHuman activity recognition using wearable sensors by deep convolutional neural networks,\u201d In: 23th ACM International Conference on Multimedia (13 Oct. 2015, Brisbane, Australia), Brisbane, 2015, pp. 1307\u20131310.","DOI":"10.1145\/2733373.2806333"},{"key":"2022020121510279073_j_comp-2020-0227_ref_007","unstructured":"H. Gjoreski, J. Bizjak, M. Gjoreski, M. Gams, \u201cComparing deep a classical machine learning methods for human activity recognition using wrist accelerometer,\u201d In: 25th International Joint Conference on Artificial Intelligence (9\u201315 July 2016, New York, USA), New York, 2016, pp. 1\u20137."},{"key":"2022020121510279073_j_comp-2020-0227_ref_008","unstructured":"T. T. T. Nguyen, N. D. Nguyen, \u201cExperiments on deep learning for wearable activity recognition,\u201d Southeast\u2013Asian J. Sci., vol. 5, pp. 101\u2013110, 2017."},{"key":"2022020121510279073_j_comp-2020-0227_ref_009","doi-asserted-by":"crossref","unstructured":"F. Hutter, L. Kotthoff, J. Vanschoren, Automated Machine Learning: Methods, Systems, Challenges, Springer Nature, UK, 2019.","DOI":"10.1007\/978-3-030-05318-5"},{"key":"2022020121510279073_j_comp-2020-0227_ref_010","unstructured":"N. Y. Hammerla, S. Halloran, T. Plots, \u201cDeep, convolutional, and recurrent models for human activity recognition using wearables,\u201d In: 25th International Joint Conference on Artificial Intelligence (9-15 July 2016, New York, USA), New York, 2016, pp. 1533\u20131540."},{"key":"2022020121510279073_j_comp-2020-0227_ref_011","doi-asserted-by":"crossref","unstructured":"C. A. Ronao, S. B. Cho, \u201cHuman activity recognition with smartphone sensors using deep learning neural networks,\u201d Expert. Syst. Appl., vol. 59, pp. 235\u2013244, 2016.","DOI":"10.1016\/j.eswa.2016.04.032"},{"key":"2022020121510279073_j_comp-2020-0227_ref_012","doi-asserted-by":"crossref","unstructured":"M. Zeng, L. T. Nguyen, B. Yu, O. J. Mengshoel, J. Zhu, P. Wu, et al., \u201cConvolutional neural networks for human activity recognition using mobile sensors,\u201d In: 6th International Conference on Mobile Computing and Services (6 Nov. 2014, Austin, USA), Austin, 2014, pp. 197\u2013205.","DOI":"10.4108\/icst.mobicase.2014.257786"},{"key":"2022020121510279073_j_comp-2020-0227_ref_013","doi-asserted-by":"crossref","unstructured":"S. Ha, J. M. Yun, S. Choi, \u201cMulti-modal convolutional neural networks for activity recognition,\u201d In: IEEE International Conference on Systems, Man, and Cybernetics (9-12 Oct. 2015, Hong Kong, China), Hong Kong, 2015, pp. 3017\u20133022.","DOI":"10.1109\/SMC.2015.525"},{"key":"2022020121510279073_j_comp-2020-0227_ref_014","unstructured":"M. A. Alsheikh, A. Selim, D. Niyato, L. Doyle, S. Lin, H. P. Tan, \u201cDeep activity recognition models with triaxial accelerometers,\u201d In: AAAI Workshop: Artificial Intelligence Applied to Assistive Technologies and Smart Environments (25 Jan. 2016, Phoenix, USA), Phoenix, 2016, pp. 8\u201313."},{"key":"2022020121510279073_j_comp-2020-0227_ref_015","doi-asserted-by":"crossref","unstructured":"S. M\u00fcnzner, P. Schmidt, A. Reiss, M. Hanselmann, R. Stiefelhagen, R. D\u00fcrichner, \u201cCNN-based sensor fusion techniques for multimodal human activity recognition,\u201d In: 2017 ACM International Symposium on Wearable Computers (11\u201315 Sep. 2017, Hawaii, USA), Hawaii, 2017, pp. 158\u2013165.","DOI":"10.1145\/3123021.3123046"},{"key":"2022020121510279073_j_comp-2020-0227_ref_016","unstructured":"P. Kasnesis, C. Z. Patrikakis, I. S. Venieris, \u201cPerceptionNet: A deep convolutional neural network for late sensor fusion,\u201d In: Intelligent Systems Conference (6\u20137 Sep. 2018, London, UK), London, 2018, pp. 1\u20139."},{"key":"2022020121510279073_j_comp-2020-0227_ref_017","doi-asserted-by":"crossref","unstructured":"V. S. Murahari, T. Pl\u00f6tz, \u201cOn attention models for human activity recognition,\u201d In: International Symposium on Wearable Computers (8\u201312 Oct. 2018, Singapore), Singapore, 2018, pp. 100\u2013103.","DOI":"10.1145\/3267242.3267287"},{"key":"2022020121510279073_j_comp-2020-0227_ref_018","doi-asserted-by":"crossref","unstructured":"R. Chavarriaga, H. Sagha, A. Calatroni, S. T. Digumarti, G. Tr\u00f6ster, J. R. Millan, et al., \u201cThe Opportunity challenge: A benchmark database for on-body sensor-based activity recognition,\u201d Pattern Recogn. Lett., vol. 34, pp. 2033\u20132042, 2013.","DOI":"10.1016\/j.patrec.2012.12.014"},{"key":"2022020121510279073_j_comp-2020-0227_ref_019","doi-asserted-by":"crossref","unstructured":"M. Zeng, H. Gao, T. Yu, O. J. Mengshoel, H. Langseth, I. Lane, et al., \u201cUnderstanding and improving recurrent networks for human activity recognition by continuous attention,\u201d In: International Symposium on Wearable Computers (8\u201312 Oct. 2018, Singapore), Singapore, 2018, pp. 56\u201363.","DOI":"10.1145\/3267242.3267286"},{"key":"2022020121510279073_j_comp-2020-0227_ref_020","doi-asserted-by":"crossref","unstructured":"J. Suto, S. Oniga, P. Pop-Sitar, \u201cFeature analysis to human activity recognition,\u201d Int. J. Comp. Commun., vol. 12, pp. 116\u2013130, 2017.","DOI":"10.15837\/ijccc.2017.1.2787"},{"key":"2022020121510279073_j_comp-2020-0227_ref_021","unstructured":"M. A. Nielsen, Neural Networks and Deep Learning, Determination Press, 2015, Ebook, http:\/\/neuralnetworksanddeeplearning.com\/, Accessed 18 January 2020."},{"key":"2022020121510279073_j_comp-2020-0227_ref_022","doi-asserted-by":"crossref","unstructured":"J. Suto, S. Oniga, \u201cEfficiency investigation of artificial neural networks in human activity recognition,\u201d J. Ambient. Intell. Human. Comput, vol. 9, pp. 1049\u20131060, 2017.","DOI":"10.1007\/s12652-017-0513-5"},{"key":"2022020121510279073_j_comp-2020-0227_ref_023","unstructured":"J. Bergstra, Y. Bengio, \u201cRandom Search for Hyper-Parameter Optimization,\u201d J. Mach. Learn. Res., vol. 13, pp. 281\u2013305, 2012."},{"key":"2022020121510279073_j_comp-2020-0227_ref_024","unstructured":"E. Brochu, V. M. Cora, N. Freitas, \u201cA tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning,\u201d arXiv:http:\/\/arXiv.org\/abs\/arXiv:1012.2599, 2010."},{"key":"2022020121510279073_j_comp-2020-0227_ref_025","doi-asserted-by":"crossref","unstructured":"C. E. Rasmussen, \u201cGaussian process in machine learning,\u201d In: Summer School on Machine Learning, Springer, Berlin, Heidelberg, 2003, pp. 63\u201371.","DOI":"10.1007\/978-3-540-28650-9_4"},{"key":"2022020121510279073_j_comp-2020-0227_ref_026","unstructured":"J. Bergstra, D. Yamins, D. D. Cox, \u201cMaking a science of model search: hyperparameter optimization in hundreds of dimensions for vision architectures,\u201d In: 30th International Conference on Machine Learning (16-21 Jun. 2013, Atlanta, USA), Atlanta, 2013, pp. 115\u2013123."},{"key":"2022020121510279073_j_comp-2020-0227_ref_027","unstructured":"J. Snoek, H. Larochelle, R. P. Adams, \u201cPractical Bayesian optimization for machine learning algorithms,\u201d In: Advances in Neural Information Processing Systems (3\u20138 Dec. 2012, Lake Tahoe, USA), Lake Tahoe, 2012, pp. 2951\u20132959."},{"key":"2022020121510279073_j_comp-2020-0227_ref_028","doi-asserted-by":"crossref","unstructured":"J. Suto, S. Oniga, \u201cEfficiency investigation from shallow to deep neural network techniques in human activity recognition,\u201d Cogn. Syst. Res., vol. 54, pp. 37\u201349, 2019.","DOI":"10.1016\/j.cogsys.2018.11.009"},{"key":"2022020121510279073_j_comp-2020-0227_ref_029","doi-asserted-by":"crossref","unstructured":"J. Suto, S. Oniga, C. Lung, I. Orha, \u201cComparison of offline and real-time human activity recognition results using machine learning techniques,\u201d Neural Comput. Appl. 32, pp. 15673\u201315686, 2018, https:\/\/doi.org\/10.1007\/s00521-018-3437-x.","DOI":"10.1007\/s00521-018-3437-x"}],"container-title":["Open Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.degruyter.com\/document\/doi\/10.1515\/comp-2020-0227\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.degruyter.com\/document\/doi\/10.1515\/comp-2020-0227\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,2,1]],"date-time":"2022-02-01T22:26:10Z","timestamp":1643754370000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.degruyter.com\/document\/doi\/10.1515\/comp-2020-0227\/html"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,1,1]]},"references-count":29,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2021,1,13]]},"published-print":{"date-parts":[[2021,1,1]]}},"alternative-id":["10.1515\/comp-2020-0227"],"URL":"https:\/\/doi.org\/10.1515\/comp-2020-0227","relation":{},"ISSN":["2299-1093"],"issn-type":[{"value":"2299-1093","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,1,1]]}}}