Prediction of Endangered Animals Using Linear Regression
S Vaishnavi, Udhaya Vani P, J Julie, J. Joshan Athanesious · 2022 1st International Conference on Computational Science and Technology (ICCST) · 2022
The paper herein will analyze and predict the population of endangered animals for the upcoming years using Machine Learning algorithms by calculating the animal's count. We develop a demographic method to appraise endangered rates and extinction probabilities from time-series data. The results can be predicted by feeding datasets which consist of parameters like population count of nearly 20 years of 30 animal species. Here, We use “LINEAR REGRESSION”, “DECISION TREE REGRESSION” and “RANDOM FOREST REGRESSOR” to obtain better predictive performance. The model's output will be the count of the animals in the next 10 years. Here we use the above-mentioned three algorithms and compare their accuracy and finally deploy the most accurate model using Stream lit. It is implemented with the latest and most popular Python 3.6. This is the first method in machine learning to find the population of animals.