Skip to content

am878/student_prediction

Repository files navigation

A Model for Success

Bashir Duale, Rafee Islam, Abhishek Mishra

Goal

Predict if students will need support after graduation based on their college applications

Data

3,700 survey questions from 16,000 students answered from 10th grade to 6 years post college graduation.

Data Processing

Student "success" defined as job-satisfaction and job alignment to career goals. Used survey responses from 12th grade students to accurately reflect college applications and narrow down data points. Used ANOVA to narrow down data to most relevant data points.

Prediction

Compared results of Linear Discriminant Analysis, Multi-Layer Perceptron, and K-Nearest Neighbors.

About

Using machine learning and data analytics to determine incoming students that will require additional support

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages