Answer Happy

◄Safari 3:42 AM Wed Jun 1 Please use your project 3 pipeline (feel free to add new models) and answer the following questions. 1. Design, execute, and report two more data mining models of your choose (Decision Trees, Gradient boosting, Forest, Neural networks) and answering the following questions. a. List of your models on the pipeline b. Explains Pros/Cons of each model Reference: Decision Tree: on-tree-algorithm-428cbd199d9a Gradient Boosting: https://towardsdatascience.com/gradient ... d-9259bd82 05af Forest: https://www.section.io/engineering-educ ... est-in-mac hine-learning/ Logistic regression: https://iq.opengenus.org/advantages-and ... egression/ Neural networks: https://subscription.packtpub.com/book/big_data_and business intelligence/978178 8397872/1/ch01lvl1sec 27/pros-and-cons-of-neural-networks 2. Assessing Models a. Run the Model Comparison node and view the results. Which model was selected? Based on what criteria? (For example: Validation Misclassification Rate was used to select the Decision Tree model) Notes: Each model should be compared and reported on one of the following accuracy measures: confusion matrix and Area under the ROC curve. Prediction Type Validation Fit Statistic Direction Decisions Misclassification smallest Average Profit/Loss largest/smallest Kolmogorov-Smirnov Statistic largest Rankings ROC Index (concordance) largest Gini Coefficient largest Estimates. smallest Average Squared Error Schwarz's Bayesian Criterion Log-Likelihood smallest largest b. Which model has the best ROC curve? Include ROC curve plots to show the performance of different models. Reference: 6%1