Open AccessImplementation Model Linear Regression, Neural Network and Support Vector Machine in Simple Mathematical Pendulum Experiments to Determine the Value of Gravity
Author(s) -
Irfan Syafar Farouk,
Mustafa Mamat,
W. S. Mada Sanjaya,
Wahid Abdurrahman,
Hani Rubiani,
Mohamad Afendee,
Wahid Abdurahman
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.b1057.0782s719
Subject(s) - artificial neural network , standard deviation , value (mathematics) , support vector machine , pendulum , computer science , mathematics , algorithm , control theory (sociology) , artificial intelligence , statistics , machine learning , engineering , mechanical engineering , control (management)
machine in mathematical pendulum experiments to find the value of gravity. There were 4 data obtained from mathematical pendulum experiments which were then interpolated to obtain more data (13 data), then the data was used as training data for each model. Each model is tested to get a gravity value of 26 including training data, then compared with reference gravity values [17,18,19]. The results of the model Neural network proved to be the most accurate with an error value of 2.53%. The support vector machine model is the most accurate model with a standard deviation value of 0.03 and the error deviation of 0.058 is the smallest value of the three models in this paper.