Bringing Transparency to Predictive Analytics: A Systematic Comparison of Predictive Modeling Methods in Higher Education
نویسندگان
چکیده
Colleges have increasingly turned to predictive analytics target at-risk students for additional support. Most of the analytic applications in higher education are proprietary, with private companies offering little transparency about their underlying models. We address this lack by systematically comparing two important dimensions: (1) different approaches sample and variable construction how these affect model accuracy (2) selection modeling approaches, ranging from methods many institutional researchers would be familiar more complex machine learning methods, affects performance stability predicted scores. The relative ranking students’ probability completing college varies substantially across approaches. While we observe substantial gains models trained on a structured represent typical enrollment spells robust set predictors, similar between simplest most
منابع مشابه
Data-intensive analytics for predictive modeling
The Data Abstraction Research Group was formed in the early 1990s, to bring focus to the work of the Mathematical Sciences Department in the emerging area of knowledge discovery and data mining (KD & DM). Most activities in this group have been performed in the technical area of predictive modeling, roughly at the intersection of machine learning, statistical modeling, and database technology. ...
متن کاملPredictive Analytics
In today's economic climate where budget reductions are common, executives are under pressure to deliver profitable growth. Business leaders must identify and implement the critical items that will enable the enterprise to remain competitive. Methods and techniques long utilized in the actuarial department are being leveraged to improve many areas of the insurance operation. Will these methods ...
متن کاملFrom Predictive to Prescriptive Analytics
Predictive analyses taking advantage of the recent explosion in the availability and accessibility of data have been made possible through flexible machine learning methodologies that are often well-suited to the variety and velocity of today’s data collection. This can be witnessed in recent works studying the predictive power of social media data and in the transformation of business practice...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: AERA Open
سال: 2021
ISSN: ['2332-8584']
DOI: https://doi.org/10.1177/23328584211037630