CFEED Machine Learning Models

How ready are your students for the next step?

CFEED’s machine learning models help schools and institutions predict readiness, reduce risk, and provide support—earlier. By analyzing longitudinal data across K–12, postsecondary, and beyond, these models surface actionable insights at critical transition points in a student’s journey.

Turning Data Into Early, Actionable Insight

Student success is shaped long before graduation day. CFEED’s machine learning models are designed to identify readiness signals and risk factors early so educators, advisors, and institutions can intervene when it matters most.

Rather than looking backward at outcomes, these models focus on:

Predicting readiness before key transitions
Identifying students who may need additional support
Informing advising, course planning, and intervention strategies

Each model addresses a specific moment in the student journey, while working together as part of a connected, longitudinal system.

Our Predictive Models

Post-secondary readiness

Predicts which students are prepared for college-level work.

The Post-Secondary Readiness model identifies academic experiences and performance patterns that signal whether a student is on track for success in higher education.

Transfer
Readiness

Measures how aligned students are with their target programs.

The Transfer Readiness model evaluates how well a student’s coursework and academic progress align with the requirements of their intended transfer institution and major.

Transfer
Shock

Identify students at risk of early disruption—and intervene sooner.

Transfer Shock focuses on the critical first terms after transition, flagging early warning signs that may indicate academic disruption allowing educators to intervene.

A Connected View of the Student Journey

Designed to Work Together

While each model addresses a specific question, they are most powerful when used together. Combined, CFEED’s machine learning models provide a longitudinal view of student readiness—from early preparation through transfer and beyond.

Institutions use these insights to:

Align coursework with long-term outcomes
Identify risk earlier across transition points
Inform policy, advising, and program design
Move from reactive reporting to proactive planning

Benefits Across the Academic Journey

From identifying readiness early to reducing attrition after transfer, CFEED’s machine learning models support student success across the full academic pathway.

Predictive readiness models identified 14,000+ dual enrollment(DE) ready students and an 11% increase at Osceola County.

Osceola County’s college-going rate increased from 47.3% to 63%.

Increased enrollment at Valencia College due to DE analytics and Osceola Prosper program by ~6,700.

Improved AA and CPP completion. 88% of students who received Helios grant for CPP completion enrolled for a second semester at UCF

Reduced attrition through transfer shock predictions. 89% of students who received part-time incentive in the 1st semester of UCF enrolled for a 2nd semester.

Improved time to degree. Students who complete 5 high impact courses see reduced attrition and time to completion.

Built for Collaboration and Action

CFEED’s models are developed and refined through multi-institutional collaboration, ensuring insights translate into real-world action—not just reports. Partners use these models to:

Deliver applied research that drives insight and impact

Test and scale evidence-based interventions

Align K-12, College, and University Strategies

Continuously improve student outcomes over time

Ready to Explore What's Possible?

CFEED’s machine learning models help institutions move from data to decisions—earlier, faster, and with greater confidence.