Stratified sampling in Cohort-based Data for Machine Learning Model Development

Notification

Announcement!

ISJEM Invites papers for various areas like engineering, Management, Science & other multi discplinary subjects. Please submit your paper for review.

ISJEM assigns a digital object identifier (DOI) to each published paper, making it easier for the paper to be cited in various major databases like Google Scholar, ResearchGate, Academia.edu, etc…

ISJEM takes 24–48 hours to publish a research paper. Within 24 hours, the submitted paper will be reviewed and notified of its status, and it will be published once the processing fee is successfully received.

Stratified sampling in Cohort-based Data for Machine Learning Model Development

Version
Download 17
File Size 357.32 KB
Download

Stratified sampling in Cohort-based Data for Machine Learning Model Development

Vaibhav Tummalapalli
Atlanta, USA
vaibhav.tummalapalli21@gmail.com

Abstract—Cohort-based data is a prevalent structure in many industries, enabling longitudinal analyses and tracking customer behaviors over time. However, sampling such data for model development presents unique challenges, especially when events (e.g., responses, purchases) are unevenly distributed across cohorts. Random sampling can introduce biases, leading to models that fail to generalize. This paper presents a stratified sampling framework designed to maintain the proportional representation of events and non-events within each cohort, even when oversampling or undersampling is applied. The approach ensures stable and unbiased models, offering insights into practical implementation and evaluation metrics

Keywords—Random, Stratified Sampling, Machine Learning, Cohort analysis, Class Imbalance, Sampling Bias.