Outlier detection in test samples and supervised training set selection | ||
| International Journal of Nonlinear Analysis and Applications | ||
| مقاله 54، دوره 12، شماره 1، مرداد 2021، صفحه 701-712 اصل مقاله (615.62 K) | ||
| نوع مقاله: Research Paper | ||
| شناسه دیجیتال (DOI): 10.22075/ijnaa.2021.4878 | ||
| نویسندگان | ||
| Navid Mohseni1؛ Hossein Nematzadeh* 2؛ Ebrahim Akbari2 | ||
| 1Department of Computer Engineering, Babol Branch, Islamic Azad University, Babol, Iran | ||
| 2Department of Computer Engineering, Sari Branch, Islamic Azad University, Sari, Iran | ||
| چکیده | ||
| Outlier detection is a technique for recognizing samples out of the main population within a data set. Outliers have negative impacts on classification. The recognized outliers are deleted to improve the classification power generally. This paper proposes a method for outlier detection in test samples besides a supervised training set selection. Training set selection is done based on the intersection of three well known similarity measures namely, jacquard, cosine, and dice. Each test sample is evaluated against the selected training set for possible outlier detection. The selected training set is used for a two-stage classification. The accuracy of classifiers are increased after outlier deletion. The majority voting function is used for further improvement of classifiers. | ||
| کلیدواژهها | ||
| Outlier detection؛ Training set selection؛ Similarity measures | ||
| مراجع | ||
|
| ||
|
آمار تعداد مشاهده مقاله: 16,235 تعداد دریافت فایل اصل مقاله: 9,663 |
||
| تعداد نشریات | 22 |
| تعداد شمارهها | 718 |
| تعداد مقالات | 10,319 |
| تعداد مشاهده مقاله | 72,323,695 |
| تعداد دریافت فایل اصل مقاله | 64,046,966 |