AIPrimer.AI
  • 🚦AI Primer In Transportation
  • CHAPTER 1 - INTRODUCTION TO MACHINE LEARNING
    • Machine Learning in Transportation
    • What is Machine Learning?
    • Types of Machine Learning
      • Supervised Learning
      • Unsupervised Learning
      • Semi-supervised Learning
      • Reinforced Learning
    • Fundamental concepts of machine learning
      • Model Training and Testing
      • Evaluating the Model’s Prediction Accuracy
      • The Underfitting and Overfitting Problems
      • Bias-Variance Tradeoff in Overfitting
      • Model Validation Techniques
      • Hyperparameter Tuning
      • Model Regularization
      • The Curse of Ddimensionality
    • Machine Learning versus Statistics
  • CHAPTER 2 - SUPERVISED METHODS
    • Supervised Learning_Complete Draft
    • K-Nearest Neighbor (KNN) Algorithm
    • Tree-Based Methods
    • Boosting
    • Support Vector Machines (SVMs)
  • CHAPTER 3 - UNSUPERVISED LEARNING
    • Principal Component Analysis
      • How Does It Work?
      • Interpretation of PCA result
      • Applications in Transportation
    • CLUSTERING
      • K-MEANS
      • SPECTRAL CLUSTERING
      • Hierarchical Clustering
    • REFERENCE
  • CHAPTER 4 - NEURAL NETWORK
    • The Basic Paradigm: Multilayer Perceptron
    • Regression and Classification Problems with Neural Networks
    • Advanced Topologies
      • Modular Network
      • Coactive Neuro–Fuzzy Inference System
      • Recurrent Neural Networks
      • Jordan-Elman Network
      • Time-Lagged Feed-Forward Network
      • Deep Neural Networks
  • CHAPTER 5 - DEEP LEARNING
    • Convolutional Neural Networks
      • Introduction
      • Convolution Operation
      • Typical Layer Structure
      • Parameters and Hyperparameters
      • Summary of Key Features
      • Training of CNN
      • Transfer Learning
    • Recurrent Neural Networks
      • Introduction
      • Long Short-Term Memory Neural Network
      • Application in transportation
    • Recent Development
      • AlexNet, ZFNet, VggNet, and GoogLeNet
      • ResNet
      • U-Net: Full Convolutional Network
      • R-CNN, Fast R-CNN, and Faster R-CNN
      • Mask R-CNN
      • SSD and YOLO
      • RetinaNet
      • MobileNets
      • Deformable Convolution Networks
      • CenterNet
      • Exemplar Applications in Transportation
    • Reference
  • CHAPTER 6 - REINFORCEMENT LEARNING
    • Introduction
    • Reinforcement Learning Algorithms
    • Model-free v.s. Model-based Reinforcement Learning
    • Applications of Reinforcement Learning to Transportation and Traffic Engineering
    • REFERENCE
  • CHAPTER 7 - IMPLEMENTING ML AND COMPUTATIONAL REQUIREMENTS
    • Data Pipeline for Machine Learning
      • Introduction
      • Problem Definition
      • Data Ingestion
      • Data Preparation
      • Data Segregation
      • Model Training
      • Model Deployment
      • Performance Monitoring
    • Implementation Tools: The Machine Learning Ecosystem
      • Machine Learning Framework
      • Data Ingestion tools
      • Databases
      • Programming Languages
      • Visualization Tools
    • Cloud Computing
      • Types and Services
    • High-Performance Computing
      • Deployment on-premise vs on-cloud
      • Case Study: Data-driven approach for the implementation of Variable Speed Limit
      • Conclusion
  • CHAPTER 8 - RESOURCES
    • Mathematics and Statistics
    • Programming, languages, and software
    • Machine learning environments
    • Tools of the Trade
    • Online Learning Sites
    • Key Math Concepts
  • REFERENCES
  • IMPROVEMENT BACKLOG
Powered by GitBook
On this page
  1. CHAPTER 1 - INTRODUCTION TO MACHINE LEARNING
  2. Types of Machine Learning

Supervised Learning

Supervised learning is one of the most common machine learning methods that is widely studied in academia and used in industry. Supervised learning is a category of machine learning methods that maps the functional relationship, between a set of inputs X and outputs y, where y and C being the number of classes. The goal of supervised learning is to estimate the function based on labeled training dataset, and then the relationship can be used to predict future outputs given a new set of unlabeled inputs. It is called supervised learning because the model is estimated based on known labeled outputs in the training dataset. In comparison with unsupervised learning, the model outcome is often unknown and not observable from training dataset. The model building process includes problem formulation, training dataset collection, feature extraction, model fitting, model evaluation, and deployment. The following chart described the supervised learning model training process at high-level.

Types of Supervised learning Algorithm

The supervised learning is a regression problem if the output is continuous and a classification problem if the output is categorical. The classification problem can be binary classification if number of outcomes C = 2, or multiclass classification problem if C > 2. Supervised learning method is often an effective way to train a function if training data can be relatively easily collected. However, collecting training dataset can often be expensive or sometimes impractical. The quality of training dataset often dictates the model predictability and accuracy. Supervised learning also makes a critical assumption that the distribution of training examples is the same as testing samples, which is often violated in real-world applications. There are many different supervised learning algorithms. The choice of different models is often determined by problem formulation, variables types (continuous vs. discrete), number of predicted outcomes (binary vs. multiclass), model training (online vs. batch), and model interpretability.

Generalized Linear Regression Models

  • Linear Regression

  • Logistic Regression

  • Multinomial Logistic Regression

  • Support Vector Machine

  • Decision Tree

  • Naïve Bayes

  • Linear Discriminant Analysis

  • K-Nearest Neighbor

  • Neural Network

Examples of Supervised Learning:

Supervised learning is one of the most widely used machine learning methods. Some of the successful implementation including the following use cases:

  • Email Spam Filtering: Email can be classified into spam and non-spam emails based on the frequency for some of the key words used in the email, which is often processed by the technique called bag of words. The algorithm is based on training dataset that already labeled the email as spam or non-spam email. The algorithm is widely implemented in email spam filtering.

  • Crash Severity Modeling: Traffic crash data is a popular data source in transportation safety and crash outcomes can often be classified into different severity levels (KABCO Injury Classification). Researchers often used multinomial logistic regression model to identify the relationship between contributing factors and crash severity outcomes.

  • Image Classification in hand writing recognition and autonomous vehicle Object Detection: One of the common data sources in transportation is video image data collected from infrastructure and autonomous vehicles. We often need to classify types of vehicles pass through an intersection and whether there are pedestrians on the sidewalk. Convolutional neural network is often used to fit the labeled images and detect the objects when presented new images.

PreviousTypes of Machine LearningNextUnsupervised Learning

Last updated 1 year ago