1 / 11

Navigating Machine Learning Algorithms for Predictive Modeling in Truck Maintenance

Dive into the world of supervised and unsupervised machine learning algorithms to predict truck failures and maintenance costs for Navistar. Explore regression and classification models such as linear regression, random forest, and k-means clustering. Discover how to select predictors, tune hyperparameters, preprocess data, and evaluate model performance to optimize predictions.

urendes
Download Presentation

Navigating Machine Learning Algorithms for Predictive Modeling in Truck Maintenance

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. STARTER TOOLBOX: MACHINE LEARNING ALGORITHMS

  2. SUPERVISED LEARNING • Predictors + Outcomes • Classification: Binary Outcome • Regression: Continuous Outcome

  3. SUPERVISED LEARNING MODELS Regression Classification • Linear Regression • Polynomial Regression • LASSO •Random Forest •Support Vector Machine •Neural Network •K Nearest Neighbor • Logistic Regression

  4. UNSUPERVISED LEARNING • Only have predictors (no outcome) • Examples: • Principal Components Analysis • K Means Clustering

  5. R PACKAGES AND FUNCTIONS Algorithm Package Main Function Linear Regression stats lm Polynomial Regression stats lm Lasso, Ridge glmnet or caret glmnet or train Random Forest randomForest randomForest Neural Net neuralnet neuralnet Support Vector Machine e1071 svm K Nearest Neighbor class knn Logistic Regression stats glm Principal Components stats prcomp K Means Clustering stats kmeans

  6. MORE THOROUGH TOOLBOX…

  7. SAMPLE OBJECTIVES FOR NAVISTAR PROJECT • Classification: • Predict if a truck will fail • Predict if truck will cost Navistar more than the mean repairs • … • Regression: • Predict when a truck will fail • Predict how much a truck will cost Navistar • … • Unsupervised learning: • Cluster trucks based on mileage, features, etc. • …

  8. OBJECTIVES FOR TODAY 1. Choose an outcome variable of interest 2. Choose a few variables you think may be meaningful predictors. 3. Choose one or two algorithms which could be used for your model 4. Research a little about these models 1. Any assumptions necessary? 2. Any hyperparameters to be chosen/tuned? 3. Any extra data manipulation necessary (for example, normalizing)

  9. OBJECTIVES FOR TODAY CONT. 5. Split data into two disjoint pieces: Training (70%) and testing (30%) 6. Fit the model on the training data set 7. Predict outcomes on the testing data set 8. Calculate the Root Mean Squared Error

  10. SOME THINGS TO THINK ABOUT

More Related