Data Mining for Scientific Applications

A large volume of complex, multi-dimensional scientific data is collected and stored daily. Data mining and predictive modeling offer a means of analysis of that data. Data mining and predictive modeling are capable of automatic extraction of knowledge deeply hidden in data, enabling discovery of knowledge not otherwise attainable.

This class is a shorter, less in-depth version of Fundamentals of Data Mining, customized for the world of science. Obtain an overview of the methods, techniques, and processes of data mining, with an emphasis on scientific applications. Explore a variety of scientific case studies learn how data mining can be applied to make meaningful conclusions, predictions, and classification of data. This course is application-focused and does not require prior programming experience.

Topics include:

• Introduction to knowledge discovery process and standards
• Classification and prediction
• Preparing input and output for data mining
• Decision tables, decision trees, classification rules
• Mining association rules
• Numeric prediction and regression
• Clustering methods
• Evaluating, testing, and verifying models

Practical experience:

• Hands-on data mining exercises

Software: WEKA is used for class assignments.

Course typically offered: Online in Winter and Summer

Prerequisites: Biostatistics, Introduction to Statistics, or Statistics for Data Analytics or equivalent knowledge required.

Next Steps: Upon completion of this course, consider taking Fundamentals of Data Mining to continue learning.

More Information: For more information about this course, please contact unex-techdata@ucsd.edu.