Impact of Climatic Change on Agricultural Product Yield Using K-Means and Multiple Linear Regressions

Abstract

Adequate information about climate change helps farmers to prepare and helps boost crop yield. Over the years, crops prediction was performed by manually considering farmer's experience on the particular crop in relation to the weather. This method was Inadequate, depends on the farmer's unreliable memory and grossly inaccurate. There is a need to introduce computational means to study and predict optimal climatic factors for improved crop growth and yield. The aim of this research work is to study the impact of climatic changes on the yield production of roots and tubers crops. K-means classification algorithm, Multiple Linear Regression, Python programming language, Flask Framework, Python machine learning packages numpy, matplotlib, Scikit-learn are the methodology used. While the obtained results show that CO2 Emission and Temperature does not really play a key role on how climate impact yield of root and tubers, rainfall plays more role; therefore, the study concludes that the three variables (temperature, rainfall, and CO2 Emission) are not enough to predict agricultural yield. It is therefore recommended that further research should be carried out to determine how other climatic factors such as soil type; humidity, sunlight etc. affect the yield of crops. The objective of this research is to study climatic change using data mining techniques, to design a predictive model using multiple linear regression to find the most optimal temperature and rainfall for effective crop yield and to simulate the multiple linear regression model design that achieve a high accuracy and a high generality in terms of climate change to crop yield.

Метод Оцінки Коефіцієнтів При Лінійних Членах Багатовимірної Поліноміальної Регресії, Заданої Надлишковим Описом

“…В багатьох інформаційних експертних системах чи в інформаційних системах класифікації використовуються багатовимірні лінійні чи нелінійні регресії, коефіцієнти яких при змінних знаходяться методом найменших квадратів [1][2][3][4][5][6][7][8]. В роботах [9][10][11] була сформульована наступна проблема.…”

Construction of a Multivariate Polynomial Given by a Redundant Description in Stochastic and Deterministic Formulations Using an Active Experiment

“…Their reliable design is still a non-trivial scientific and applied problem. In particular, the problem of constructing a multivariate polynomial regression (MPR) is still the subject of research by scientists [1][2][3][4][5][6][7][8]. Papers [9,10] addressed the following problem.…”

“…This allows for arbitrary 𝑐 < 𝑑 to use at the given 𝑛 only one set of NOPFs 𝑄 𝑗 (𝑧), 𝑗 = 0, 𝑟, built for the values of the scalar variable 𝑧 (8). Indeed [12], let the NOPF be 𝑄 𝑗 (𝑧) = 𝑞 𝑗0 + 𝑞 𝑗1 𝑧 + ⋯ + 𝑞 𝑗𝑗 𝑧 𝑗 , 𝑗 = 0, 𝑟, where the numbers 𝑞 𝑖𝑗 , 𝑗 = 0, 𝑟, are obtained by numbers 𝑧 𝑗 , 𝑗 = 1, 𝑛 (see (8)). Then the estimates 𝛾 ̂𝑗, 𝑗 = 0, 𝑟, obtained by the least squares method (LSM) according to the results of a virtual experiment 𝑧 𝑖 → 𝑦 𝑖 , 𝑖 = 1, 𝑛 (𝑦 𝑖 , 𝑖 = 1, 𝑛, are the output results of an active experiment for a UPR (2) for 𝑥 𝑖 , 𝑖 = 1, 𝑛, which are related to the values of 𝑧 𝑖 , 𝑖 = 1, 𝑛, by the expressions ( 8), ( 9)), have the form…”

Risk Forecasting of Data Confidentiality Breach Using Linear Regression Algorithm

“…The analysis of different areas of information security management in organizations of various activities is not left out: the management of information security system resources in the organization [19][20][21]; risk management and threats to information security [20,22,23]; management of documentation and information assistance system in the organization [19,20,24,25]; information security audit management [23,26,27]; management of information security system efficiency analysis [19,21,28]. Some sources are associated with the use of different forecasting methods based on the construction of appropriate models [29][30][31][32][33][34][35].…”