Machine Learning: Algorithms in the Real World

This course introduces professionals to problem definition and data preparation in the context of machine learning projects. Learners will gain the ability to clearly define a machine learning problem, survey available data resources, and prepare data for effective machine learning applications. The course sets the foundation for the subsequent modules in the specialization.

Learners will delve into supervised learning techniques, exploring real case studies and analyzing business scenarios. They will gain practical skills in implementing decision trees, k-nearest neighbors, and support vector machines, while also learning to address common production issues in applied machine learning. Prior knowledge of Python programming and basic understanding of linear algebra and statistics are recommended for this course.

Participants will focus on understanding the critical elements of data in the learning, training, and operation phases of applied machine learning. The course equips learners with the skills to mitigate biases, improve model generality, tackle overfitting, and enhance model accuracy through thoughtful feature engineering. Additionally, they will explore the impact of algorithm parameters on model strength.

This final course synthesizes the knowledge gained throughout the specialization, guiding learners through a complete machine learning project and the development of a maintenance roadmap. Participants will learn to deal with changing data, identify potential unintended effects, and define procedures for the operationalization and maintenance of applied machine learning models.