Determining the order of a polynomial model for construction of trend lines in data science problems

Authors

DOI:

https://doi.org/10.18372/2073-4751.71.17001

Keywords:

data science, models for building a trend line, polynomial model

Abstract

The work deals with the problem of improving data science technologies, which are now widely used in many industries. The quality of the implementation of these technologies is largely determined by the accuracy of the calculation of trend dependence parameters, which requires an adequate determination of the order of the polynomial model. The purpose of the work is to improve the methods of determining the order of the polynomial model for constructing a trend line in data science tasks.

The authors proposed an approach to determining the order of a polynomial model for building a trend line in data science tasks, which is based on the analysis of the values of higher derivatives of the experimental curve, taking into account measurement errors. The results of evaluating the effectiveness of the proposed approach are given.

References

David Dietrich. Data Science & Big Data Analytics: Discovering, Analyzing, Visualizing and Presenting Data / David Dietrich, Barry Heller, Beibei Yang. – John Wiley & Sons, Inc., Indianapolis, Indiana, 2015. – 420 p.

Sage Andrew, Melsa James, Estimation Theory With Applications to Communications and Control. – McGraw-Hill Book Company, Inc.; First Edition, 1971. – 752 p.

Published

2022-11-01

Issue

Section

Статті