150 words to each question
Q1 Linear Regression (LR) is one of the simplest machine learning algorithms that comes under Supervised Learning technique and used for solving regression problems (Linear Regression vs Logistic Regression, n.d.). It is used for predicting the continuous dependent variable with the help of independent variables. The goal of LR is to find the best fit line that can predict the output for the continuous dependent variable (Mondal, 2020). If predictions are made using a single independent variable, then it is called Simple Linear Regression and if multiple independent variables are used then is called Multiple Linear Regression. By finding the best fit line, we can establish a relationship between the dependent and independent variable. The relationship should be of linear nature. The output for LR should only be a continuous value such as price, age, salary, etc. Figure 1 shows LR.
Figure 1. Linear Regression. This figure is an example of simple linear regression.
Logistic regression is one of the most popular machine learning algorithms that comes under supervised learning techniques (Linear Regression vs Logistic Regression, n.d.). Although it is mainly used for classification, it can also be used for regression problems. Logistic regression is used to predict the categorical dependent variable with the help of independent variables. The output can be only between the 0 and 1 (Mondal, 2020). This approach is used where the probabilities between two classes is required, for example, if we want to know if it will rain today or not (either 0 or 1, true or false, etc.). Logistic regression is based on the concept of Maximum Likelihood estimation, where the observed data should be most probable. Figure 2 shows a logistic regression curve.
Logistic Regression Curve
Figure 2. Logistic Regression Curve. This figure is an example of a logistic regression curve.
Linear Regression vs Logistic Regression. (n.d.). Retrieved from Java T Point: https://www.javatpoint.com/linear-regression-vs-logistic-regression-in-machine-learning
Mondal, S. (2020, December 1). Beginners Take: How Logistic Regression is Related to Linear Regression. Retrieved from Analytics Vidhya: https://www.analyticsvidhya.com/blog/2020/12/beginners-take-how-logistic-regression-is-related-to-linear-regression/
Q2. 1. Discuss the difference between linear regression and logistic regression.
A: Linear Regression: Linear regressions follow a line within a quartile of probability either in a negative regression or a positive regression. “The outcome variable is continuous” (EMC Education Services, 2015, p. 222). The assumption must be made that the input and outcome have some kind of relationship following a line when scatter plotted. Even when the relationships do not have obvious relationships, linear regressions should still be considered, and changes to the data may need to be made in order for them to fit into the parameters of the evaluation.
Logistic Regression: Logistic regressions are most often used when the dependent variable is categorical (Swaminathan, 2018, p. np). The output is often in the form of probabilities such as: there is an 80% that the phone call will be spam. There are three main types of logistic regression: Binary regressions, multinomial regressions, ordinal regressions. Binary regressions only have two possible outcomes; True or False. Multinomial regressions have three or more categories without a specific ordering; Dodge, Ford, Honda. Ordinal regression contains three or more categories that have ordering; 1-star to 5-stars. One potential use for a logistic regression could the classification of new species of animals, given physical traits through perhaps a decision tree.
Difference & Similarities: Linear regression is used to answer a regression question while logistic regressions are used to handle the classification of different types of problems. Linear regression finds the best fit line to explain the data, while logistic regression works to fit the data to the curve (Mondal, 2020, p. np). Both regressions us supervised machine learning using titled datasets to train an algorithm to either classify or predict data/outcomes, while using linear equations for predictions.
EMC Education Services. (2015). Data Science & Big Data Analytics: Discovering, Analyzing, Visualizing and Presenting Data. Indianapolis, IN: John Wiley & Sons, Inc.
Mondal, S. (2020, December 1). Beginners Take: How Logistic Regression is related to Linear Regression. Retrieved from Analytics Vidhya: https://www.analyticsvidhya.com/blog/2020/12/beginners-take-how-logistic-regression-is-related-to-linear-regression/
Swaminathan, S. (2018, March 15). Logistic Regression – Detailed Overview. Retrieved from Towards Data Science : https://towardsdatascience.com/logistic-regression-detailed-overview-46c4da4303bc