Open AccessOutlier Detection of the Agricultural Time Series
Author(s) -
I.L. Vorotnikov,
A. G. Rozanov,
Maria Leonidovna Sidelnikova,
Sergei Tkachev,
L Volochuk
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/723/4/042070
Subject(s) - outlier , series (stratigraphy) , anomaly detection , reliability (semiconductor) , computer science , time series , field (mathematics) , normal distribution , distribution (mathematics) , probability distribution , data mining , statistics , mathematics , artificial intelligence , machine learning , paleontology , mathematical analysis , power (physics) , physics , quantum mechanics , pure mathematics , biology
To increase the reliability of the results of processing the data of automated monitoring of the dynamics of time series in the agricultural field of activity, it is necessary to assess the probability distribution of the data flow. It is shown that, if the distribution deviates from the normal one, adaptive and robust procedures for identifying and correcting random outliers should be applied.