Text Document Clustering: Wordnet vs. TF-IDF vs. Word Embeddings
Michał Marcińczuk, Mateusz Gniewkowski, Tomasz Walkowiak, Marcin Będkowski
Abstract
In the paper, we deal with the problem of unsupervised text document clustering for the Polish language. Our goal is to compare the modern approaches based on language modeling (doc2vec and BERT) with the classical ones, i.e., TF-IDF and wordnet-based. The experiments are conducted on three datasets containing qualification descriptions. The experiments’ results showed that wordnet-based similarity measures could compete and even outperform modern embedding-based approaches.- Anthology ID:
- 2021.gwc-1.24
- Volume:
- Proceedings of the 11th Global Wordnet Conference
- Month:
- January
- Year:
- 2021
- Address:
- University of South Africa (UNISA)
- Editors:
- Piek Vossen, Christiane Fellbaum
- Venue:
- GWC
- SIG:
- SIGLEX
- Publisher:
- Global Wordnet Association
- Note:
- Pages:
- 207–214
- Language:
- URL:
- https://aclanthology.org/2021.gwc-1.24
- DOI:
- Cite (ACL):
- Michał Marcińczuk, Mateusz Gniewkowski, Tomasz Walkowiak, and Marcin Będkowski. 2021. Text Document Clustering: Wordnet vs. TF-IDF vs. Word Embeddings. In Proceedings of the 11th Global Wordnet Conference, pages 207–214, University of South Africa (UNISA). Global Wordnet Association.
- Cite (Informal):
- Text Document Clustering: Wordnet vs. TF-IDF vs. Word Embeddings (Marcińczuk et al., GWC 2021)
- PDF:
- https://preview.aclanthology.org/corrections-2024-05/2021.gwc-1.24.pdf