Outlier Detection Using the Multiobjective Genetic Algorithm
Agnieszka Duraj, Łukasz Chomątek · Wydawnictwo Politechniki Łódzkiej · 2020
Since almost all datasets may be affected by the presence of anomalies which may skew the interpretation of data, outlier detection has become a crucial element of many datamining applications. Despite the fact that several methods of outlier detection have been proposed in the literature, there is still a need to look for new, more effective ones. This paper presents a new approach to outlier identification based on genetic algorithms. The study evaluates the performance and examines the features of several multiobjective genetic algorithms.