Yıl: 2023 Cilt: 31 Sayı: 5 Sayfa Aralığı: 751 - 770 Metin Dili: İngilizce DOI: 10.55730/1300-0632.4016 İndeks Tarihi: 22-11-2023

Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification

Öz:
Although the standard k-nearest neighbor (KNN) algorithm has been used widely for classification in many different fields, it suffers from various limitations that abate its classification ability, such as being influenced by the distribution of instances, ignoring distances between the test instance and its neighbors during classification, and building a single/weak learner. This paper proposes a novel algorithm, called stepwise dynamic nearest neighbor (SDNN), which can effectively handle these problems. Instead of using a fixed parameter k like KNN, it uses a dynamic neighborhood strategy according to the data distribution and implements a new voting mechanism, called stepwise voting. Experimental results were conducted on 50 benchmark datasets. The results showed that the proposed SDNN method outperformed the KNN method, KNN variants, and the state-of-the-art methods in terms of accuracy.
Anahtar Kelime: Machine learning classification k-nearest neighbor majority voting ensemble learning

Belge Türü: Makale Makale Türü: Araştırma Makalesi Erişim Türü: Erişime Açık
0
0
0
  • [1] García-Pedrajas N, Del Castillo JAR, Cerruela-García G. A proposal for local k values for k-nearest neighbor rule. IEEE Transactions on neurol networks and learning systems 2017; 28 (2): 470-475. https://doi.org/10.1109/TNNLS.2015.2506821
  • [2] Lu J, Qian W, Li S, Cui R. Enhanced k-nearest neighbor for intelligent fault diagnosis of rotating machinery. Applied Science 2021; 11 (3): 919-934. https://doi.org/ 10.3390/app11030919
  • [3] Gou J, Ma H, Ou W, Zeng S, Rao Y et al. A generalized mean distance-based k-nearest neighbor classifier. Expert Systems with Applications 2019; 115: 356-372. https://doi.org/10.1016/j.eswa.2018.08.021
  • [4] Saadatfar H, Khosravi S, Joloudari JH, Mosavi A, Shamshirband S. A new k-nearest neighbors classifier for big data based on efficient data pruning. Mathematics 2020; 8 (2): 1-12. https://doi.org/10.3390/math8020286
  • [5] Durbin M, Wonders MA, Flaska M, Lintereur A. K-nearest neighbors regression for the discrimination of gamma rays and neutrons in organic scintillators. Nuclear Instruments and Methods in Physics Research 2021; 987: 1-9. https://doi.org/10.1016/j.nima.2020.164826
  • [6] Xia J, Zhang J, Wang Y, Han L, Yan H. WC-KNNG-PC: Watershed clustering based on k-nearest-neighbor graph and Pauta Criterion. Pattern Recognition 2022; 121: 1-14. https://doi.org/10.1016/j.patcog.2021.108177
  • [7] Yu L, Yue W, Wang S, Lai KK. Support vector machine based multiagent ensemble learning for credit risk evaluation. Expert Systems with Applications 2010; 37 (2): 1351-1360. https://doi.org/10.1016/j.eswa.2009.06.083
  • [8] Okediran TM, Vincent OR, Abayomi-Alli AA, Adeniaran OJ. An IOT-based COVID-19 detector using k-nearest neighbor. Decision Sciences for COVID-19, International Series in Operations Research & Management Science 2022; 320: 27-43. https://doi.org/10.1007/978-3-030-87019-5_2
  • [9] Timuş O, Doğru Bolat E. K-NN-based classification of sleep apnea types using ECG. Turkish Journal of Electrical Engineering and Computer Sciences 2017; 25 (4): 3008-3023. https://doi.org/10.3906/elk-1511-99
  • [10] Dilmaç S, Ölmez Z, Ölmez T. Comparative analysis of MABC with KNN, SOM, and ACO algorithms for ECG heartbeat classification. Turkish Journal of Electrical Engineering and Computer Sciences 2018; 26 (6): 2819-2830. https://doi.org/10.3906/elk-1712-328
  • [11] Tang H, Xu Y, Lin A, Heidari AA, Wang M et al. Predicting green consumption behaviors of students using efficient firefly grey wolf-assisted k-nearest neighbor classifiers. IEEE Access 2020; 8: 35546-35562. https://doi.org/10.1109/ACCESS.2020.2973763
  • [12] Hemachandran K, George PM, Rodriguez R, Kulkarni RM, Roy S. Performance analysis of k-nearest neighbor classification algorithms for bank loan sectors. Smart Intelligent Computing and Communication Technology 2020; 38: 9-13. https://doi.org/10.3233/APC210004
  • [13] Qureshi S, Karrila S, Vanichayobon S. Human sleep scoring based on k-nearest neighbors. Turkish Journal of Electrical Engineering and Computer Sciences 2018; 26 (6): 2802-2818. https://doi.org/10.3906/elk-1805-12
  • [14] Dilmaç F, Alpkoçak A. Automatic keyword assignment system for medical research articles using nearest- neighbor searches. Turkish Journal of Electrical Engineering and Computer Sciences 2022; 30 (5): 1821–1838. https://doi.org/10.55730/1300-0632.3907
  • [15] Wolff B, Lorenz E, Kramer O. Statistical learning for short-term photovoltaic power predictions. Computational Sustainability 2016; 645: 31–45. https://doi.org/10.1007/978-3-319-31858-5_3
  • [16] Geler Z, Kurbalija V, Ivanovic M, Radovanovic M. Weighted kNN and constrained elastic distances for time-series classification. Expert System with Applications 2020; 162: 1-12. https://doi.org/10.1016/j.eswa.2020.113829
  • [17] Kang S. K-nearest neighbor learning with graph neural networks. Mathematics 2021; 9 (8): 1-12. https://doi.org/10.3390/math9080830
  • [18] Wu J, Cai Z, Gao Z. Dynamic k-nearest-neighbor with distance and attribute weighted for classification. In: International Conference on Electronics and Information Engineering; Kyoto, Japan; 2010. pp. 356-360.
  • [19] Zhong X-F, Guo S-Z, Gao L, Shan H, Zheng J-H. An improved k-NN classification with dynamic k. In: International Conference on Machine Learning and Computing; Singapore; 2017. pp. 211–216.
  • [20] Witten IH, Frank E, Hall MA, Pal CJ. Data Mining: Practical Machine Learning Tools and Techniques. 4th ed. Cambridge, MA, USA: Morgan Kaufmann, 2016.
  • [21] Liu Y. Bsnsing: a decision tree induction method based on recursive optimal boolean rule composition. Informs Journal on Computing 2022; 34 (6): 2867-3350. https://doi.org/10.1287/ijoc.2022.1225
  • [22] Tomita TM, Browne J, Shen C, Chung J, Patsolic JL et al. Sparse projection oblique randomer forests. Journal of Machine Learning Research 2020; 21 (1): 4193–4231.
  • [23] Abellan J, Mantas CJ, Castellano JG. AdaptativeCC4.5: Credal C4.5 with a rough class noise estimator. Expert Systems with Applications 2018; 92: 363–379. https://doi.org/10.1016/j.eswa.2017.09.057
APA KARABAŞ D, Birant D, YILDIRIM TAŞER P (2023). Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. , 751 - 770. 10.55730/1300-0632.4016
Chicago KARABAŞ Deniz,Birant Derya,YILDIRIM TAŞER Pelin Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. (2023): 751 - 770. 10.55730/1300-0632.4016
MLA KARABAŞ Deniz,Birant Derya,YILDIRIM TAŞER Pelin Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. , 2023, ss.751 - 770. 10.55730/1300-0632.4016
AMA KARABAŞ D,Birant D,YILDIRIM TAŞER P Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. . 2023; 751 - 770. 10.55730/1300-0632.4016
Vancouver KARABAŞ D,Birant D,YILDIRIM TAŞER P Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. . 2023; 751 - 770. 10.55730/1300-0632.4016
IEEE KARABAŞ D,Birant D,YILDIRIM TAŞER P "Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification." , ss.751 - 770, 2023. 10.55730/1300-0632.4016
ISNAD KARABAŞ, Deniz vd. "Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification". (2023), 751-770. https://doi.org/10.55730/1300-0632.4016
APA KARABAŞ D, Birant D, YILDIRIM TAŞER P (2023). Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. Turkish Journal of Electrical Engineering and Computer Sciences, 31(5), 751 - 770. 10.55730/1300-0632.4016
Chicago KARABAŞ Deniz,Birant Derya,YILDIRIM TAŞER Pelin Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. Turkish Journal of Electrical Engineering and Computer Sciences 31, no.5 (2023): 751 - 770. 10.55730/1300-0632.4016
MLA KARABAŞ Deniz,Birant Derya,YILDIRIM TAŞER Pelin Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. Turkish Journal of Electrical Engineering and Computer Sciences, vol.31, no.5, 2023, ss.751 - 770. 10.55730/1300-0632.4016
AMA KARABAŞ D,Birant D,YILDIRIM TAŞER P Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. Turkish Journal of Electrical Engineering and Computer Sciences. 2023; 31(5): 751 - 770. 10.55730/1300-0632.4016
Vancouver KARABAŞ D,Birant D,YILDIRIM TAŞER P Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification. Turkish Journal of Electrical Engineering and Computer Sciences. 2023; 31(5): 751 - 770. 10.55730/1300-0632.4016
IEEE KARABAŞ D,Birant D,YILDIRIM TAŞER P "Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification." Turkish Journal of Electrical Engineering and Computer Sciences, 31, ss.751 - 770, 2023. 10.55730/1300-0632.4016
ISNAD KARABAŞ, Deniz vd. "Stepwise dynamic nearest neighbor (SDNN): a new algorithm for classification". Turkish Journal of Electrical Engineering and Computer Sciences 31/5 (2023), 751-770. https://doi.org/10.55730/1300-0632.4016