Identification of Outliers Based on the Sensitivity of Data
Author(s):
Prof. Shiva Sumanth Reddy, Amogh S Bharadwaj, Chandrakanth N Murthy, H Vishwajit, Harsh V Challa
Keywords:
Data Preprocessing ( Anova Test ) Data Analytics, Dataset Sensitivity, Box plot, Isolation Forest, Local Outlier Factor.
Abstract
Outlier detection is examined and can be applied in many fields and domains. The Outliers occur because of the manual errors which are most often made by humans while data entry, fraudulent error when a malicious practice occurs, system behaviour, by natural deviations in datasets or instrument error. The detection of outliers has been used over many years, to identify outliers and analyse outliers where required. Sometimes the outliers need to be removed, Sometimes we use it to identify Outliers. The main challenge in outlier detection is to work on data that is highly Sensitive data. Sometimes the data is so sensitive that the outlier data coincides with the normal data, this usually occurs in the domain of malicious activities. The proposed system assists to clean data in less time and great accuracy. Our paper focuses on outlier detection which can be applied to various domains that have time series data. This shows critical review on various approaches to detect outliers and provides the most accurate technique for particular type of data.
Article Details
Unique Paper ID: 159743
Publication Volume & Issue: Volume 9, Issue 12
Page(s): 339 - 347
Article Preview & Download
Share This Article
Conference Alert
NCSST-2021
AICTE Sponsored National Conference on Smart Systems and Technologies
Last Date: 25th November 2021
SWEC- Management
LATEST INNOVATION’S AND FUTURE TRENDS IN MANAGEMENT