CampusLink: An ML-Based Campus Event Discovery and Recommendation System

Author

Kabilan V, Samuel Santhosam S, Poojitha V, Dr. G. Paavai Anand

Abstract

 Campus event announcements distributed across messaging groups and social-media pages are difficult to search and personalize. This paper presents CampusLink, a campus event discovery and registration system that integrates text classification, semantic retrieval and hybrid recommendation. A React interface communicates with a FastAPI backend. TF-IDF with Logistic Regression assists organizers in assigning event categories; keyword retrieval and a pretrained MiniLM encoder support discovery; reciprocal rank fusion combines their rankings. A recommender integrates student interests, participation similarity, popularity and negative feedback. Evaluation separates authored examples, synthetic events and reviewed campus announcements. Classification achieves 78.6% accuracy and a macro-F1 of 0.793 on 28 reviewed real announcements after training on 72 authored examples. On a 200-event synthetic retrieval benchmark, TF-IDF achieves nDCG@5 of 0.817 on the topical and constrained portion of a 45-query test split. A recommendation experiment with 30 simulated users yields Recall@10 of 0.167 for the hybrid method and 0.100 for popularity-only ranking. Semantic and fusion performance remain unmeasured on the larger and real-event benchmarks. The results establish a functioning applied ML prototype and identify the need for independently labelled data and real-user evaluation.

Keywords

Campus events; machine learning; TF-IDF; semantic search; reciprocal rank fusion; hybrid recommendation; event classification.

Full Text:

Download Paper PDF

References

[1] X. Liu, Q. He, Y. Tian, W.-C. Lee, J. McPherson, and J. Han, “Event-based social networks: Linking the online and offline social worlds,” Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1032–1040, 2012. doi: 10.1145/2339530.2339693.

[2] G. Salton and C. Buckley, “Term-weighting approaches in automatic text retrieval,” Information Processing & Management, vol. 24, no. 5, pp. 513–523, 1988. doi: 10.1016/0306-4573(88)90021-0.

[3] F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011. https://jmlr.org/papers/v12/pedregosa11a.html

[4] N. Reimers and I. Gurevych, “Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks,” Proceedings of EMNLP-IJCNLP, pp. 3982–3992, 2019. doi: 10.18653/v1/D19-1410.

[5] Sentence Transformers, “all-MiniLM-L6-v2,” model card. https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2 (accessed October 30, 2026).

[6] G. V. Cormack, C. L. A. Clarke, and S. Büttcher, “Reciprocal rank fusion outperforms condorcet and individual rank learning methods,” Proceedings of the 32nd International ACM SIGIR Conference, pp. 758–759, 2009. doi: 10.1145/1571941.1572114.

[7] P. Resnick, N. Iacovou, M. Suchak, P. Bergstrom, and J. Riedl, “GroupLens: An open architecture for collaborative filtering of netnews,” Proceedings of the ACM Conference on Computer Supported Cooperative Work, pp. 175–186, 1994. doi: 10.1145/192844.192905.

[8] K. Järvelin and J. Kekäläinen, “Cumulated gain-based evaluation of IR techniques,” ACM Transactions on Information Systems, vol. 20, no. 4, pp. 422–446, 2002. doi: 10.1145/582415.582418.

[9] Scikit-learn developers, “Common pitfalls and recommended practices,” scikit-learn documentation. https://scikit-learn.org/stable/common_pitfalls.html (accessed October 30, 2026).

Share your valuable work from Social Media Buttons