# Regression Analysis In Machine Learning

732 Words3 Pages
Review paper – Regression Analysis in Machine Learning Abstract- Machine learning is the indispensable quality of artificial intelligence. Predicting or numeric featuring is called Regression in the statistical literature and it is the subject of research in both machine learning and statistics. Regression analysis in its simplest form is a method of fitting of a line through a given set of data points plotted on a graph in which the points are dependent on both the parameters which are stated on the x-axis and y-axis. Through regression analysis we aim to build predictive models which will consist of parameters on which we input the value of one parameter and get the output of the other parameter (no matter which technique of regression we…show more content…
One of the main features of supervised learning algorithms is that they model dependencies and relationships between the target output and input features to predict the value for new data. Regression algorithms predict the output values based on input features from the data fed in the system. The go-to methodology is the algorithm builds a model on the features of Types of Regression- There are 7 main types of regression models which are used in machine learning and data interpretation Linear regression- Linear regression which is of two types simple and multiple is the first of the few common things taught in learning the predictive model, it creates the relation between two types of dependent and independent where the independent can be of two types that is continuous and discrete. This type of regression is used to calculate the value of the outcome of the variable Y based on the input of the variable X. The mathematical equation of the linear regression is as follows: Y= β1 + β2X + ϵ Where β1 is the intercept and β2 is the slope. They are called the regression coefficients and ϵ is the error term Linear Regression can be represent by the equation of line y=(a+b)*x +c where a and b are the intercept and c is the error term given in the equation we have different ways to find out the values of a and…show more content…
Then we have to remove collinearity which is followed by gaussian distributions and then comes rescaling the inputs. To obtain the best fit line- the least square line method is the easiest and most common way of making a regression line. If the datasets are well defined there is no better regression than linear regression, but it can suffer from multicollinearity, heteroscedasticity etc. Logistic regression- The logistic regression model is the method to fit a regression curve of y and x where Y=f(x) when y is categorical variable.it is used to find the probability of how much chance there is suh a case that the event is success or the same event is a failure. Logistic regression model is binary to check is if the answer is 0 or 1 as in on or off ; true or false. This model is also called as the binomial logistic model since the variable to predict is binary, also this model can also predict the data series if there is a dependent variable which can assume more than 2