Course Outline
Introduction
- Differentiation between statistical learning (statistical analysis) and machine learning
- Adoption of machine learning technology and talent by financial and banking institutions for government applications
Different Types of Machine Learning
- Supervised learning versus unsupervised learning
- Iteration and evaluation processes
- Bias-variance trade-off in model selection
- Combining supervised and unsupervised learning techniques (semi-supervised learning)
Machine Learning Languages and Toolsets
- Comparison of open-source versus proprietary systems and software for government use
- Evaluation of Python, R, and Matlab in the context of governmental needs
- Overview of essential libraries and frameworks for machine learning tasks
Machine Learning Case Studies
- Utilization of consumer data and big data analytics for government operations
- Assessing risk in consumer and business lending within the public sector
- Enhancing customer service through sentiment analysis for government services
- Detecting identity fraud, billing fraud, and money laundering in governmental systems
Hands-on: Python for Machine Learning
- Preparing the development environment for machine learning projects
- Obtaining necessary Python machine learning libraries and packages
- Working with scikit-learn and PyBrain in government applications
How to Load Machine Learning Data
- Data sources including databases, data warehouses, and streaming data for government use
- Distributed storage and processing using Hadoop and Spark for large-scale governmental datasets
- Exported data and Excel integration in government systems
Modeling Business Decisions with Supervised Learning
- Classifying data (classification) for informed decision-making in government
- Predictive modeling using regression analysis for government outcomes
- Selecting appropriate machine learning algorithms for governmental needs
- Understanding and applying decision tree algorithms in public sector contexts
- Exploring random forest algorithms for enhanced accuracy in government models
- Evaluating model performance to ensure reliability for government applications
- Practical exercise for hands-on experience
Regression Analysis
- Linear regression techniques for government data analysis
- Generalizations and handling nonlinearity in governmental datasets
- Practical exercise to reinforce learning
Classification
- Review of Bayesian principles relevant to government applications
- Naive Bayes classification for efficient data categorization in the public sector
- Logistic regression for binary outcomes in governmental datasets
- K-Nearest neighbors for proximity-based classification in government systems
- Practical exercise to apply classification methods
Hands-on: Building an Estimation Model
- Assessing lending risk based on customer type and history for government financial services
Evaluating the Performance of Machine Learning Algorithms
- Cross-validation and resampling techniques to ensure model reliability in government applications
- Bootstrap aggregation (bagging) for improved predictive accuracy in public sector models
- Practical exercise to evaluate algorithm performance
Modeling Business Decisions with Unsupervised Learning
- Addressing scenarios where sample data sets are limited in government contexts
- K-means clustering for data segmentation in public sector operations
- Challenges and considerations of unsupervised learning in governmental applications
- Exploring advanced techniques beyond K-means, including Bayes networks and Markov Hidden Models
- Practical exercise to apply unsupervised learning methods
Hands-on: Building a Recommendation System
- Analyzing past customer behavior to improve new service offerings in government programs
Extending Your Company's Capabilities
- Developing machine learning models in the cloud for scalable government solutions
- Accelerating machine learning processes with GPU technology for enhanced performance in government operations
- Applying deep learning neural networks for advanced tasks such as computer vision, voice recognition, and text analysis in governmental applications
Closing Remarks
Requirements
- Experience with Python programming for government applications
- Basic familiarity with statistics and linear algebra for government data analysis
Testimonials (2)
the ML ecosystem not only MLFlow but Optuna, hyperops, docker , docker-compose
Guillaume GAUTIER - OLEA MEDICAL
Course - MLflow
I enjoyed participating in the Kubeflow training, which was held remotely. This training allowed me to consolidate my knowledge for AWS services, K8s, all the devOps tools around Kubeflow which are the necessary bases to properly tackle the subject. I wanted to thank Malawski Marcin for his patience and professionalism for training and advice on best practices. Malawski approaches the subject from different angles, different deployment tools Ansible, EKS kubectl, Terraform. Now I am definitely convinced that I am going into the right field of application.
