{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T06:52:35Z","timestamp":1777704755536,"version":"3.51.4"},"reference-count":33,"publisher":"SAGE Publications","issue":"1","license":[{"start":{"date-parts":[[2020,5,4]],"date-time":"2020-05-04T00:00:00Z","timestamp":1588550400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Journal of Intelligent &amp; Fuzzy Systems"],"published-print":{"date-parts":[[2020,7,17]]},"abstract":"<jats:p>One of the momentous transformation performed by an artificial neural network (ANN), Support Vector Machine (SVM), Radial basis Function (RBF) and many other machine learning method is the application of activation function. MyAct the proposed activation method is used here with various ANN architectures for link prediction, classification and general prediction. Statistical properties of data used here to prove the effectiveness of proposed activation function MyAct over other popular activation methods. A data dependent transfer method is developed, which is pioneer in its own way. This proves to be an unified formulation for the robust and generalised learning for the classification, link prediction and regression problem types. Classification is done with Iris dataset using ANN with different activation method and results are compared. Improved results are achieved when MyAct used with Tailored Deep Feed Forward Artificial Neural Network (TDFFANN), simple Artificial Neural Network and Deep Artificial Neural Network.<\/jats:p>\n                  <jats:p>Aim here is to develop a novel activation method which work with positive data, negative data, small size data, big size data, skewed data or corrupt data. An attempt is made to cover complete versatile behaviour of data. Currently not a single activation method can work well on all above mentioned data. Results obtained using MyAct on the datasets used here proves it to be a good choice in comparison to logsig, tansig and other popular activation methods for classification and link prediction.<\/jats:p>\n                  <jats:p>Satisfactory improvement is achieved by using data length as well as negative range values in the prediction done by proposed method. MyAct had 22% better standard deviation than ReLU (Rectified Linear unit) and 36. 28% better standard deviation than ELU (Exponential linear unit). MyAct has 2. 6% better accuracy in regression error than Swiss method and 2. 5% better accuracy in regression error than ELU. Other results are discussed in the paper.<\/jats:p>","DOI":"10.3233\/jifs-191618","type":"journal-article","created":{"date-parts":[[2020,5,5]],"date-time":"2020-05-05T10:51:02Z","timestamp":1588675862000},"page":"665-677","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":2,"title":["Assessment of effectiveness of data dependent activation method: MyAct"],"prefix":"10.1177","volume":"39","author":[{"given":"Sandhya","family":"Pundhir","sequence":"first","affiliation":[{"name":"USICT, GGSIPU, Dwarka, Delhi, India"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Udayan","family":"Ghose","sequence":"additional","affiliation":[{"name":"USICT, GGSIPU, Dwarka, Delhi, India"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Upasana","family":"Bisht","sequence":"additional","affiliation":[{"name":"KIT, Pitampura, Delhi, India"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","published-online":{"date-parts":[[2020,5,4]]},"reference":[{"key":"e_1_3_1_2_2","unstructured":"RamachandranP. ZophB. and LeQ.V. \u201cSwish: a self-gated activation function. \u201carXiv preprint arXiv:1710. 05941. (2017)."},{"key":"e_1_3_1_3_2","unstructured":"Carlile Brad et al. \u201cImproving deep learning by inverse square root linear units (ISRLUs).\u201d arXiv preprint arXiv:1710. 09967 (2017)."},{"key":"e_1_3_1_4_2","unstructured":"Farhadi Farnoush Partovi NiaVahid and LodiAndrea \u201cActivation adaptation in neural networks.\u201d arXiv preprint arXiv:1901. 09849 (2019)."},{"key":"e_1_3_1_5_2","doi-asserted-by":"crossref","unstructured":"Laudani Antonino et al. \u201cOn training efficiency and computational costs of a feed forward neural network:Areview.\u201d Computational intelligence and neuroscience 2015 (2015).","DOI":"10.1155\/2015\/818243"},{"key":"e_1_3_1_6_2","doi-asserted-by":"publisher","DOI":"10.1162\/NECO_a_00849"},{"key":"e_1_3_1_7_2","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-010-0407-3"},{"key":"e_1_3_1_8_2","unstructured":"Koturwar Saiprasad and MerchantShabbir \u201cWeight initialization of deep neural networks (DNNs) using data statistics.\u201d arXiv preprint arXiv:1710. 10570 (2017)."},{"key":"e_1_3_1_9_2","unstructured":"Izmailov Pavel et al. \u201cAveraging weights leads to wider optima and better generalization.\u201d arXiv preprint arXiv:1803. 05407 (2018)."},{"key":"e_1_3_1_10_2","doi-asserted-by":"crossref","unstructured":"GodfreyL.B. and GashlerM.S. \u201cA continuum among logarithmic linear and exponential functions and its potential to improve generalization in neural networks\u201d. In: Knowledge Discovery Knowledge Engineering and Knowledge Management (IC3K) 2015 7th International Joint Conference on 1 (2015) 481\u2013486. IEEE.","DOI":"10.5220\/0005635804810486"},{"key":"e_1_3_1_11_2","unstructured":"Alcantara Giovanni \u201cEmpirical analysis of non-linear activation functions for Deep Neural Networks in classification tasks.\u201d arXiv preprint arXiv:1710. 11272 (2017)."},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.4236\/jbise.2010.36083"},{"key":"e_1_3_1_13_2","unstructured":"Pedamonti Dabal \u201cComparison of non-linear activation functions for deep neural networks on MNIST classification task.\u201d arXiv preprint arXiv:1804. 02763 (2018)."},{"key":"e_1_3_1_14_2","unstructured":"Clevert Djork-Arn\u00e9 Unterthiner and HochreiterSepp \u201cFast and accurate deep network learning by exponential linear units (elus).\u201d arXiv preprint arXiv:1511. 07289 (2015)."},{"key":"e_1_3_1_15_2","unstructured":"Wang Bao et al. \u201cDeep neural networks with data dependent implicit activation function.\u201d (2018)."},{"key":"e_1_3_1_16_2","unstructured":"Morcos AriS. et al. \u201cOn the importance of single directions for generalization.\u201d arXiv preprint arXiv:1803. 06959 (2018)."},{"key":"e_1_3_1_17_2","doi-asserted-by":"crossref","unstructured":"Sandhya Ghose Udayan \u201csan sim: Factual and efficient URL text similarity algorithm.\u201d 2017 3rd International Conference on Applied and Theoretical Computing and Communication Technology (iCATccT). IEEE 2017.","DOI":"10.1109\/ICATCCT.2017.8389161"},{"key":"e_1_3_1_18_2","unstructured":"Sandhya Ghose Udayan and BishtUpasana \u201cTailored feedforward artificial neural network based link prediction.\u201d International Journal of Information Technology (2019) 1\u20139."},{"key":"e_1_3_1_19_2","unstructured":"https:\/\/archive.ics.uci.edu\/ml\/datasets\/Iris Last Accessed on 09. 09. 2019"},{"key":"e_1_3_1_20_2","unstructured":"http:\/\/www-personal.umich.edu\/\u223cmejn\/netdata\/ Last Accessed on 09. 09. 2019"},{"key":"e_1_3_1_21_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10115-014-0789-0"},{"key":"e_1_3_1_22_2","doi-asserted-by":"crossref","unstructured":"Gao Fei et al. \u201cLink prediction methods and their accuracy for different social networks and network metrics \u201d Scientific programming 2015 (2015).","DOI":"10.1155\/2015\/172879"},{"key":"e_1_3_1_23_2","unstructured":"Saha Snehanshu et al. \u201cEvolution of novel activation functions in neural network training with applications to classification of exoplanets.\u201d arXiv preprint arXiv:1906.01975 (2019)."},{"key":"e_1_3_1_24_2","doi-asserted-by":"crossref","unstructured":"Akuzawa Kei IwasawaYusuke and MatsuoYutaka \u201cAdversarial invariant feature learning with accuracy constraint for domain generalization.\u201d arXiv preprint arXiv:1904. 12543 (2019).","DOI":"10.1007\/978-3-030-46147-8_19"},{"key":"e_1_3_1_25_2","doi-asserted-by":"crossref","unstructured":"ChengC.H. HuangC.H. RuessH. and YasuokaH. \u201cTowards dependability metrics for neural networks. \u201cIn 2018 16th ACM\/IEEE International Conference on Formal Methods and Models for System Design (MEM-OCODE) on 2018 Oct 15 (pp. 1\u20134). IEEE.","DOI":"10.1109\/MEMCOD.2018.8556962"},{"key":"e_1_3_1_26_2","unstructured":"Ghauch Hadi et al. \u201cA Unified Framework for Training Neural Networks.\u201d arXiv preprint arXiv:1805. 09214 (2018)."},{"key":"e_1_3_1_27_2","unstructured":"Zhong Kai et al. \u201cRecovery guarantees for one-hidden-layer neural networks.\u201d Proceedings of the 34th International Conference on Machine Learning-Volume 70. JMLR. org 2017."},{"key":"e_1_3_1_28_2","unstructured":"Wang Xuezhi et al. \u201cPractical Compositional Fairness: Understanding Fairness in Multi-Task ML Systems.\u201d arXiv preprint arXiv:1911. 01916 (2019)."},{"key":"e_1_3_1_29_2","unstructured":"Fromherz Aymeric et al. \u201cFast Geometric Projections for Local Robustness Certification.\u201d arXiv preprint arXiv:2002. 04742 (2020)."},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.3390\/electronics8080832"},{"key":"e_1_3_1_31_2","unstructured":"Panigrahi Abhishek ShettyAbhishek and GoyalNavin \u201cEffect of Activation Functions on the Training of Overparametrized Neural Nets.\u201d arXiv preprint arXiv:1908. 05660 (2019)."},{"key":"e_1_3_1_32_2","unstructured":"https:\/\/en.wikipedia.org\/wiki\/Learning_curve Last accessed on 16. 02. 2020."},{"key":"e_1_3_1_33_2","unstructured":"He Fengxiang WangBohan and TaoDacheng \u201cPiecewise linear activations substantially shape the loss surfaces of neural networks.\u201d arXiv preprint arXiv:2003. 12236 (2020)."},{"key":"e_1_3_1_34_2","unstructured":"Nicolae Andrei \u201cPLU: The piecewise linear unit activation function.\u201d arXiv preprint arXiv:1809. 09534 (2018)."}],"container-title":["Journal of Intelligent &amp; Fuzzy Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/JIFS-191618","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.3233\/JIFS-191618","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/JIFS-191618","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T09:41:56Z","timestamp":1777455716000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.3233\/JIFS-191618"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,5,4]]},"references-count":33,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,7,17]]}},"alternative-id":["10.3233\/JIFS-191618"],"URL":"https:\/\/doi.org\/10.3233\/jifs-191618","relation":{},"ISSN":["1064-1246","1875-8967"],"issn-type":[{"value":"1064-1246","type":"print"},{"value":"1875-8967","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,5,4]]}}}