Impact of Dimensionality ReductionOn Outlier Detection Techniques
عنوان مقاله: Impact of Dimensionality ReductionOn Outlier Detection Techniques
شناسه ملی مقاله: SASTECH07_105
منتشر شده در هفتمین سمپوزیوم بین المللی پیشرفتهای علوم و تکنولوژی در سال 1391
شناسه ملی مقاله: SASTECH07_105
منتشر شده در هفتمین سمپوزیوم بین المللی پیشرفتهای علوم و تکنولوژی در سال 1391
مشخصات نویسندگان مقاله:
M.M Tavakoli - MSc. CSE Department of Shiraz University, Kerman, Iran
Ashkan Sami - Assistant Professor CSE Department of Shiraz University, Shiraz, Iran
خلاصه مقاله:
M.M Tavakoli - MSc. CSE Department of Shiraz University, Kerman, Iran
Ashkan Sami - Assistant Professor CSE Department of Shiraz University, Shiraz, Iran
In statistics, an outlier is an observation that is distant from the rest of the data. In other word mining or detecting outliers is referred to sequence of operations that lead to find exceptional objects that deviate from the rest of the data set. In this paper we use a correlation based feature selection algorithm, for dimensionality reduction as a preprocessing phase to outlier detection algorithms, which causes improvement on results of different types of outlier detection techniques.
کلمات کلیدی: Outlier Detection, Dimensionality Reduction, Unsupervised Learning
صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/205241/