This project contains the steps to analyze the characteristics of animals and the classifications of animals in 7 different categories. To classify animals in 7 different categories we need the best machine learning model which can classify animals efficiently. In this project we compared 5 different machine learning models to see which model has the best accuracy score Here used K-Nearest Neighbors, Support Vector Machine, Decision tree, Random Forest and Perceptron models. Trained them with the same data set and compared their accuracy.
zoo.csv and class.csv is used as a training data. The zoo.csv data contains 18 variables and 101 rows. These animals are classified in 7 catagories and this are available in class.csv.
Data is available at: https://www.kaggle.com/uciml/zoo-animal-classification/data
To run this project, you need the python jupiter notebook. python packages: numpy, pandas, matplotlib, os, seaborn, scikit-learn, also pydotplus for decision tree graph.
The Random Forest model has the best accuracy score and 0.97 of cross validation.
https://drive.google.com/drive/folders/1kaP3XCaoWgKiZBExzwEHLWaUMdBh4oFM?usp=sharing