It will cover the concepts and techniques that underlie current educational “student success” and “early warning” systems, giving you insight into how learners are categorized as at-risk through automated processes.
You will gain hands-on experience building these kinds of predictive models using the popular (and free) Weka software package. Also, included in this course is a discussion of supervised machine learning techniques, feature selection, model fit, and evaluation of data based on student attributes. Throughout the course, the ethical and administrative considerations of educational predictive models will be addressed.
What you'll learn
- How to use the Weka toolkit to analyze educational data and make predictions about student outcomes
- Techniques underlying supervised machine learning, including decision trees and naïve Bayes modeling
- How to apply feature selection to identify relevant attributes in the data
- How to rigorously evaluate educational predictive models
- The state of the practice in current generation educational predictive models
Syllabus
Week 1: Prediction
- Predictive models vs. explanatory models
- The predictive modeling lifecycle
- Predictive models of student success
- Ethical considerations with predictive models
- Overview of the state of the practice in educational predictive models
Week 2: Supervised Learning
- Supervised machine learning techniques, including Decision Trees and Naive Bayes
Week 3: Model Evaluation
- Making predictions
- Model evaluation and comparison
- Practical considerations
Prerequisites
We highly recommend that you take the previous course in this series before beginning this course:
Cluster Analysis
This course is intended for those who have a bachelor’s degree and are interested in developing learning and data science skills for employment in education, corporate, nonprofit, and military sectors. Experience with programming and statistics will be beneficial to participants.