TY  - JOUR
T1  - A Weighted Similarity Measure for k-Nearest Neighbors Algorithm
AU  - Karabulut, Bergen
AU  - Arslan, Güvenç
AU  - Ünver, Halil Murat
PY  - 2019
DA  - December
DO  - 10.18466/cbayarfbe.618964
JF  - Celal Bayar University Journal of Science
JO  - CBUJOS
PB  - Manisa Celal Bayar University
WT  - DergiPark
SN  - 1305-130X
SP  - 393
EP  - 400
VL  - 15
IS  - 4
LA  - en
AB  - One of the most important problems in machinelearning, which has gained importance in recent years, is classification. Thek-nearest neighbors (kNN) algorithm is widely used in classification problembecause it is a simple and effective method. However, there are several factorsaffecting the performance of kNN algorithm. One of them is determining anappropriate proximity (distance or similarity) measure. Although the Euclideandistance is often used as a proximity measure in the application of the kNN,studies show that the use of different proximity measures can improve the performanceof the kNN. In this study, we propose the Weighted Similarity k-NearestNeighbors algorithm (WS-kNN) which use a weightedsimilarity as proximity measure in the kNN algorithm. Firstly, itcalculates the weight of each attribute and similarity between the instances inthe dataset. And then, it weights similarities by attribute weights and createsa weighted similarity matrix to use as proximity measure. The proposedalgorithm is compared with the classical kNN method based on the Euclideandistance. To verify the performance of our algorithm, experiments are made on10 different real-life datasets from the UCI (UC Irvine Machine LearningRepository) by classification accuracy. Experimental results show that theproposed WS-kNN algorithm can achieve comparative classification accuracy. Forsome datasets, this new algorithm gives highly good results. In addition, we demonstratedthat the use of different proximity measures can affect the classificationaccuracy of kNN algorithm.
KW  - Classification
KW  - Feature Weighting
KW  - k-Nearest Neighbors
KW  - Weighted Similarity
CR  - 1.	Jordan, MI, Mitchell, TM. 2015. Machine learning: Trends, perspectives, and prospects. Science; 349(6245): 255-260.
CR  - 2.	Singh, A, Thakur, N, Sharma, A. A review of supervised machine learning algorithms. In 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom), IEEE, March 2016, pp. 1310-1315.
CR  - 3.	Cekik, R, Telceken, S. 2018. A new classification method based on rough sets theory. Soft Computing; 22(6): 1881-1889.
CR  - 4.	Soofi, AA, Awan, A. 2017. Classification Techniques in Machine Learning: Applications and Issues. Journal of Basic and Applied Sciences; 13: 459-465.
CR  - 5.	Aggarwal, CC. 2014. Instance-Based Learning: A Survey. Data Classification: Algorithms and Applications, 157.
CR  - 6.	Angiulli, F, Narvaez, E. 2018. Pruning strategies for nearest neighbors competence preservation learners. Neurocomputing; 308: 8-20.
CR  - 7.	Prasath, VB, Alfeilat, HAA, Lasassmeh, O, Hassanat, A. 2017. Distance and Similarity Measures Effect on the Performance of K-Nearest Neighbors Classifier-A Review. arXiv preprint arXiv:1708.04321.
CR  - 8.	Lei, Y, Zuo, MJ. 2009. Gear crack level identification based on weighted K nearest neighbors classification algorithm. Mechanical Systems and Signal Processing; 23(5): 1535-1547.
CR  - 9.	Khateeb, N, Usman, M. Efficient Heart Disease Prediction System using K-Nearest Neighbors Classification Technique, In Proceedings of the International Conference on Big Data and Internet of Thing ACM, December 2017, pp. 21-26.
CR  - 10.	Li, Q, Li, W, Zhang, J, Xu, Z. 2018. An improved k-nearest-neighbors method to diagnose breast cancer. Analyst; 143(12): 2807-2811.
CR  - 11.	Liu, Y, Wang, X, Yan, K. 2018. Hand gesture recognition based on concentric circular scan lines and weighted K-nearest neighbors algorithm. Multimedia Tools and Applications; 77(1): 209-223.
CR  - 12.	Rodrigues, ÉO. 2018. Combining Minkowski and Cheyshev: New distance proposal and survey of distance metrics using k-nearest neighbours classifier. Pattern Recognition Letters; 110: 66-71.
CR  - 13.	Mulak, P, Talhar, N. 2015. Analysis of Distance Measures Using K-Nearest Neighbors Algorithm on KDD Dataset. International Journal of Science and Research; 4(7): 2101-2104.
CR  - 14.	Hu, LY, Huang, MW, Ke, SW, Tsai, CF. 2016. The distance function effect on k-nearest neighbors classification for medical datasets. SpringerPlus; 5(1): 1304.
CR  - 15.	Dialameh, M, Jahromi, MZ. 2017. A general feature-weighting function for classification problems. Expert Systems with Applications; 72: 177-188.
CR  - 16.	Jiao, L, Pan, Q, Feng, X, Yang, F. An evidential k-nearest neighbors classification method with weighted attributes, In Proceedings of the 16th International Conference on Information Fusion, IEEE, July 2013, pp. 145-150.
CR  - 17.	Marchiori, E. Class dependent feature weighting and k-nearest neighbors classification, In IAPR International Conference on Pattern Recognition in Bioinformatics, Springer, Berlin, Heidelberg, 2013, June, pp. 69-78.
CR  - 18.	Hassanat, AB. 2014. Dimensionality invariant similarity measure. Journal of American Science; 10(8).
CR  - 19.	Alkasassbeh, M, Altarawneh, GA, Hassanat, A. 2015. On enhancing the performance of nearest neighbour classifiers using hassanat distance metric. Canadian Journal of Pure and Applied Sciences (CJPAS); 9(1).
CR  - 20.	Chomboon, K, Chujai, P, Teerarassamee, P, Kerdprasop, K, Kerdprasop, N. An empirical study of distance metrics for k-nearest neighbors algorithm, In Proceedings of the 3rd International Conference on Industrial Application Engineering, March 2015.
CR  - 21.	Kayaalp, N, Arslan, G. 2014. A Fuzzy Bayesian Classifier with Learned Mahalanobis Distance. International Journal of Intelligent Systems; 29(8): 713-726.
CR  - 22.	Kayaalp, N. Arslan, G. A New Fuzzy Bayesian Classification Approach, The 4th International Fuzzy Systems Symposium, İstanbul, 5-6 November 2015.
CR  - 23.	Greco, S, Matarazzo, B, Slowinski, R. 2001. Rough sets theory for multicriteria decision analysis. European journal of operational research; 129(1): 1-47.
UR  - https://doi.org/10.18466/cbayarfbe.618964
L1  - https://dergipark.org.tr/en/download/article-file/904927
ER  -