2 hours lecture/3 hours laboratory
3 semester credit hours/meets part of advanced informatics competencies
The increased digitization of biomedical data has dramatically increased interest in methods to analyze large quantities of data. Data mining is the process of transforming this raw data into actionable knowledge, which has led to many spectacular advances in biomedicine. This course provides an introduction to data mining methods from a biomedical perspective. The primary focus will be on practical and commonly used machine learning techniques for data mining (e.g., decision trees, support vector machines, clustering) and how these techniques transform data into knowledge. Students will engage in hands-on projects that expose them to data mining methods. Further, students will be able to critically evaluate the appropriateness of data mining methods on different tasks.
This course is designed to accommodate students with a varying degree of technical skills. No programming experience is required.
After taking BMI 6323, students will be able to:
· Describe common data mining algorithms, including their inputs, outputs, and how they work.
· Distinguish between data mining methods in terms of their appropriateness for a given type of data.
· Apply several data mining methods to diverse types of biomedical data.
· Evaluate the results of a data mining method on a given task, including identifying possible flaws in reported results.