International Journal of Electrical and Computer Engineering (IJECE) Vol. 12, No. 1, February 2022, pp. 770~775 ISSN: 2088-8708, DOI: 10.11591/ijece.v12i1.pp770-775  770 Journal homepage: http://ijece.iaescore.com Solving multiple linear regression problem using artificial neural network Mohammad S. Khrisat, Ziad A. Alqadi Computer Engineering Department, Faculty of Engineering Technology, Applied University, Amman, Jordan Article Info ABSTRACT Article history: Received Nov 14, 2020 Revised Jul 13, 2021 Accepted Jul 26, 2021 Multiple linear regressions are an important tool used to find the relationship between a set of variables used in various scientific experiments. In this article we are going to introduce a simple method of solving a multiple rectilinear regressions (MLR) problem that uses an artificial neural network to find the accurate and expected output from MLR problem. Different artificial neural network (ANN) types with different architecture will be tested, the error between the target outputs and the calculated ANN outputs will be investigated. A recommendation of using a certain type of ANN based on the experimental results will be raised. Keywords: Artificial neural network Convolutional artificial neural network Feedforward artificial neural network Multiple linear regression This is an open access article under the CC BY-SA license. Corresponding Author: Mohammad S. Khrisat Computer Engineering Department, Faculty of Engineering Technology, Applied University Amman 11134, P.O. Box 15008 Jordan Email: mkhrisat@bau.edu.jo 1. INTRODUCTION Multiple rectilinear regressions (MLR) [1] are the foremost common kind of linear regression analysis. As a prognostic analysis, it will not justify the link between one continuous dependent variable and 2 or a lot of freelance variables. The independent variables will be continuous or categorical (dummy coded as appropriate). MLR presented in (1) and Figure 1 can be solved using MATLAB Figure 2 [2]-[4]. = 0 + 1  1 + 2  2 +⋯+    (1) Where: y = dependent variable x = explanatory variables a = constants The coefficients will be as follows: a={5.0, 2.0, 3.5}, So MLR problem can be represented using (2).  = 5 + 2 1 + 3.5 2 (2) Artificial neural network (ANN) [5]-[7] is a computational mathematical model, which contains a set of fully connected neurons, which are arranged in two or more layers. Each neuron as shown in Figure 3 implements two operations [8]-[10]: i) finding the sum of products of the inputs and the associated weights [11], [12], ii) according to the selected activation function calculate the neuron output.