Using data science for detecting outliers with k Nearest Neighbors graph

Data science is a process for extracting knowledge from data using fundamental principles of analytical techniques such as statistics in order to achieve business goals. Detecting outliers is one case of data science which try to find extreme values or odd from a set of data based on the techniques and the principles of statistical calculations where data previously not utilized being to be utilized. It is intended to improve the quality of decision making in order to achieve business goal. This study tried to do the analysis and modeling of data science for detecting outliers by using k nearest neighbors graph. Finally, this study delivers the model of data science for detecting outliers by using k Nearest Neighbors (kNN) graph with k-distance calculation method.