One of the most important problems in machine learning, which has gained importance in recent years, is classification. The knearest neighbors (kNN) algorithm is widely used in classification problem because it is a simple and effective method. However, there are several factors affecting the performance of kNN algorithm. One of them is determining an appropriate proximity (distance or similarity) measure. Although the Euclidean distance 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 performance of the kNN. In this study, we propose the Weighted Similarity kNearest Neighbors algorithm (WSkNN) which use a weighted similarity as proximity measure in the kNN algorithm. Firstly, it calculates the weight of each attribute and similarity between the instances in the dataset. And then, it weights similarities by attribute weights and creates a weighted similarity matrix to use as proximity measure. The proposed algorithm is compared with the classical kNN method based on the Euclidean distance. To verify the performance of our algorithm, experiments are made on 10 different reallife datasets from the UCI (UC Irvine Machine Learning Repository) by classification accuracy. Experimental results show that the proposed WSkNN algorithm can achieve comparative classification accuracy. For some datasets, this new algorithm gives highly good results. In addition, we demonstrated that the use of different proximity measures can affect the classification accuracy of kNN algorithm.
Birincil Dil  en 

Bölüm  Makaleler 
Yazarlar 

Tarihler 
Yayımlanma Tarihi : 30 Aralık 2019 
Bibtex  @araştırma makalesi { cbayarfbe618964,
journal = {Celal Bayar University Journal of Science},
issn = {1305130X},
eissn = {13051385},
address = {},
publisher = {Celal Bayar Üniversitesi},
year = {2019},
volume = {15},
pages = {393  400},
doi = {10.18466/cbayarfbe.618964},
title = {A Weighted Similarity Measure for kNearest Neighbors Algorithm},
key = {cite},
author = {KARABULUT, Bergen and ARSLAN, Güvenç and ÜNVER, Halil Murat}
} 
APA  KARABULUT, B , ARSLAN, G , ÜNVER, H . (2019). A Weighted Similarity Measure for kNearest Neighbors Algorithm. Celal Bayar University Journal of Science , 15 (4) , 393400 . DOI: 10.18466/cbayarfbe.618964 
MLA  KARABULUT, B , ARSLAN, G , ÜNVER, H . "A Weighted Similarity Measure for kNearest Neighbors Algorithm". Celal Bayar University Journal of Science 15 (2019 ): 393400 <https://dergipark.org.tr/tr/pub/cbayarfbe/issue/50875/618964> 
Chicago  KARABULUT, B , ARSLAN, G , ÜNVER, H . "A Weighted Similarity Measure for kNearest Neighbors Algorithm". Celal Bayar University Journal of Science 15 (2019 ): 393400 
RIS  TY  JOUR T1  A Weighted Similarity Measure for kNearest Neighbors Algorithm AU  Bergen KARABULUT , Güvenç ARSLAN , Halil Murat ÜNVER Y1  2019 PY  2019 N1  doi: 10.18466/cbayarfbe.618964 DO  10.18466/cbayarfbe.618964 T2  Celal Bayar University Journal of Science JF  Journal JO  JOR SP  393 EP  400 VL  15 IS  4 SN  1305130X13051385 M3  doi: 10.18466/cbayarfbe.618964 UR  https://doi.org/10.18466/cbayarfbe.618964 Y2  2019 ER  
EndNote  %0 Celal Bayar Üniversitesi Fen Bilimleri Dergisi A Weighted Similarity Measure for kNearest Neighbors Algorithm %A Bergen KARABULUT , Güvenç ARSLAN , Halil Murat ÜNVER %T A Weighted Similarity Measure for kNearest Neighbors Algorithm %D 2019 %J Celal Bayar University Journal of Science %P 1305130X13051385 %V 15 %N 4 %R doi: 10.18466/cbayarfbe.618964 %U 10.18466/cbayarfbe.618964 
ISNAD  KARABULUT, Bergen , ARSLAN, Güvenç , ÜNVER, Halil Murat . "A Weighted Similarity Measure for kNearest Neighbors Algorithm". Celal Bayar University Journal of Science 15 / 4 (Aralık 2020): 393400 . https://doi.org/10.18466/cbayarfbe.618964 
AMA  KARABULUT B , ARSLAN G , ÜNVER H . A Weighted Similarity Measure for kNearest Neighbors Algorithm. Celal Bayar Univ J Sci. 2019; 15(4): 393400. 
Vancouver  KARABULUT B , ARSLAN G , ÜNVER H . A Weighted Similarity Measure for kNearest Neighbors Algorithm. Celal Bayar University Journal of Science. 2019; 15(4): 400393. 