2026-10-13T12:07:53Z https://data.mendeley.com/oai

oai:data.mendeley.com/sffnc5dxgv.1 2026-08-06T13:00:45Z

HOSSIN, MD Codeforces User–Problem Interaction Dataset for Personalized Programming Problem Recommendation and Balanced Team Formation Mendeley Data This dataset contains processed user–problem interaction data collected from the publicly available Codeforces API to support research on personalized competitive programming problem recommendation and balanced team formation. The recommendation dataset consists of interactions from 1,000 Codeforces users, 10,864 programming problems, and approximately 918,763 processed user–problem interactions. It includes user submission histories, problem metadata, user performance features, and user weakness profiles computed across programming tags and difficulty levels. The data were generated through a preprocessing pipeline that aggregates multiple submissions into user–problem interactions, computes solving statistics, and extracts problem ratings and programming tags. For the team formation experiments, an additional dataset was collected from 30 members of the Comilla University Programming Club, whose publicly available Codeforces submission histories were retrieved through the Codeforces API to construct user performance profiles and evaluate the proposed balanced team formation method. This dataset is intended for research in recommender systems, collaborative filtering, educational data mining, competitive programming analytics, and automated team formation. It accompanies the research titled A Hybrid Recommendation and Team Formation System for Competitive Programming and is provided to support reproducibility and future research. Data Mining Machine Learning Educational Interest Recommendation System Rahman, Md. Zahidur Dataset doi:10.17632/sffnc5dxgv.1 oai:data.mendeley.com/sffnc5dxgv.1 Creative Commons Attribution 4.0 International http://creativecommons.org/licenses/by/4.0 https://data.mendeley.com/datasets/sffnc5dxgv 2026-08-06T13:00:45Z