EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Hybrid TF-IDF and Embedding Model for Improving Similarity and Clustering Accuracy

Ero Wahyu Pratomo, Ema Utami

JOINTECS (Journal of Information Technology and Computer Science) · 2025

DOI: 10.31328/jointecs.v10i1.7344 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JOINTECS (Journal of Information Technology and Computer Science), 2025. Verfasst von Ero Wahyu Pratomo und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study addresses improving accuracy in final project document recommendation systems for students by employing ahybrid approach that combines the vector-based TF-IDF method with semantic embedding models SBERT and E5. Theprimary issue tackled is the limitation of TF-IDF in capturing semantic context, often resulting in recommendations that areless relevant in meaning. This research aims to develop a recommendation system that considers literal word similarity andsemantic similarity between documents. The Research utilizes 500 metadata records of student theses from an institutionalrepository via the OAI-PMH protocol, focusing on key attributes such as title, abstract, and keywords. The data undergoes atext preprocessing phase, which includes tokenization, normalization, stopword removal, and stemming. Three similarityapproaches are applied: TF-IDF, SBERT, and E5, each used to calculate document similarity. Evaluation compares similarityscores, analyzes ranking shifts, score distributions, and document clustering using the K-Means algorithm. Experimentalresults show that the hybrid model significantly improves recommendation quality. The E5 model produces the most stableand semantically relevant similarity scores and the cleanest and most visually distinct document clusters. Both Spearman andPearson correlation analyses indicate that combining TF-IDF with embedding models improves the ranking order ofrecommended documents. In conclusion, the hybrid approach using TF-IDF and semantic embeddings, particularly E5,significantly enhances the accuracy of document recommendation systems and the effectiveness of topic clustering

Zitieren

APA

Pratomo, E. W., & Utami, E. (2025). Hybrid TF-IDF and Embedding Model for Improving Similarity and Clustering Accuracy. JOINTECS (Journal of Information Technology and Computer Science). https://doi.org/10.31328/jointecs.v10i1.7344

Vancouver

Pratomo EW, Utami E. Hybrid TF-IDF and Embedding Model for Improving Similarity and Clustering Accuracy. JOINTECS (Journal of Information Technology and Computer Science). 2025. doi:10.31328/jointecs.v10i1.7344

BibTeX
@article{Pratomo2025,
  author = {Pratomo, Ero Wahyu and Utami, Ema},
  title = {Hybrid TF-IDF and Embedding Model for Improving Similarity and Clustering Accuracy},
  journal = {JOINTECS (Journal of Information Technology and Computer Science)},
  year = {2025},
  doi = {10.31328/jointecs.v10i1.7344}
}
Kostenlos starten →