OUTLIER DETECTION
- Version
- Download 40
- File Size 636.46 KB
- File Count 1
- Create Date 22 April 2024
- Last Updated 22 April 2024
OUTLIER DETECTION
C.Vasuki1, S.Boobathiraja2, M.Keerthana3, S.Lohit4, C.Sowmiya5
1 Assistant Professor in Information Technology, Nandha Engineering College, Perundurai-638052, Erode District, Tamilnadu, India.
2,3,4,5 Students of Information Technology, Nandha Engineering College, Perundurai-638052, Erode District, Tamilnadu, India.
ABSTRACT
To find observations that differ considerably from the bulk of the data points, outlier detection is an essential task in data analysis. To put it more simply, outliers are individual data points that stand out from the rest of the dataset. the Iris dataset, a machine learning benchmark, is used for outlier detection. The Rank SVM method, which is typically utilized for ranking jobs but has been modified for outlier identification, is used to find outliers. Standardizing the features pre-processes the dataset, which consists of measurements of iris flower sepal and petal diameters. The standardised dataset is used to train a Rank SVM model. According to the model's categorization, outliers are anticipated to be categorized as -1 and inliers as 1. The study shows the indices of outliers found in the Iris dataset and sheds light on how well Rank SVM works for outlier identification tasks.
Keywords: outlier detection, iris dataset, human behavior, machine learning
Download