Channel: Learn By Watch

Category: Education

Tags: classification algorithmsmachine learning classification pythonmachine learning coursemachine learning playlistmachine learning classificationclassification problem in machine learning examplemachine learning tutorialcoding implementation of a classification problemshow classification works in machine learningclassification in machine learningmachine learning for beginnersmachine learning full coursecoding implementation in machine learning

Description: Coding implementation of a classification problem by: 1. Read the data. 2. Create dependent and independent data sets based on our dependent and independent features. 3. Split the data into training and testing sets. 4. Train the model using different algorithms such as KNN, Decision tree, SVM, etc. 5. Evaluate the classifier. 6. Choose the classifier with the most accuracy. Here we have the car data from the ucl repository which can be found here. We need to classify the cars based on different classes. ● Extracted our features and labels from the dataset. ● Since we cannot work with string data types therefore we need to convert them into numerical format which we did using Label Encoder. ● Split our dataset into training and test dataset ● Next we apply different models on our dataset. The first one is KNN where we get an accuracy of 75% ● Then we tried fitting our dataset using Logistic regression where the accuracy was found to be 70% ● Naive Bayes’ was applied next and had an accuracy of 57.5% ● Support vector machine had an accuracy of about 74% ● Finally we applied Random Forest which gave an accuracy of 75%