Elamur Research⌕
☰
home / papers / 2026 / Towards Building a Computational Sense Inventory from the Monier-Williams Dictionary Using Clustering Techniques

Towards Building a Computational Sense Inventory from the Monier-Williams Dictionary Using Clustering Techniques

Paper · 2026 · Anagha Pradeep, Sriram Krishnan and Radhika Mamidi

Summary

Sanskrit has no complete, machine-usable sense inventory. The Sanskrit WordNet is unfinished, and the DCS provides sense labels for only part of its lexicon. This gap limits both classic NLP pipelines and LLM-based work on word sense disambiguation. The authors start from the aligned, structured Monier-Williams (MW) dictionary of Patel and Kulkarni (2024). They found that it separates senses inconsistently across parts of speech, and that roots (dhātus) and indeclinables (avyayas) were often left unsplit.

To restructure the senses, they embedded each definition with five encoders and compared them by cosine similarity. Four graph or hierarchical clustering methods then grouped the senses into synset-like clusters, making polysemy and synonymy explicit. The clusters were judged intrinsically and against a human-annotated gold sample. Louvain and Leiden proved the most robust, while HAC over-fragmented the senses. The authors describe this as the first systematic clustering-based conversion of MW into a computational sense inventory. They release it openly as a basis for WSD, MT, IR and a future Sanskrit-English WordNet.

Key points

  • Source data: The aligned MW version contains 287,604 entries and 194,072 unique lemmas, against 133,373 entries in Apte. The DCS dictionary merges senses into fewer entries (179,806).
  • Models and algorithms: The encoders were LaBSE, MPNet, Para-Mini, Qwen and Vya. The clustering methods were K-hop, HAC, Louvain and Leiden. Spectral, Affinity Propagation and hub-based clustering were also tried but performed poorly.
  • Evaluation: Louvain led on intrinsic measures for every model, while all algorithms were close on extrinsic, gold-based evaluation. The final choice gave more weight to the extrinsic results.
  • Known problems: HAC produced many singleton clusters, which inflated its separation scores. Models also struggled with metaphorical senses and with near-identical glosses. For example, hari has 31 original entries and 48 gold clusters, and its animal senses overlapped.
  • Resource: MW-Sense-Inventory is released on GitHub at three levels: the base aligned version, the annotated and experimental samples, and the full inventory.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 works

🧭 Topics 1Research subjects and languages the work is about.Word sense disambiguation · 83 works

Cite

Cite as Anagha Pradeep, Sriram Krishnan and Radhika Mamidi. 2026. Towards Building a Computational Sense Inventory from the Monier-Williams Dictionary Using Clustering Techniques. Proceedings of the 8th International Sanskrit Computational Linguistics Symposium.

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{pradeep2026building,
  title = {{Towards Building a Computational Sense Inventory from the Monier-Williams Dictionary Using Clustering Techniques}},
  author = {Anagha Pradeep, Sriram Krishnan and Radhika Mamidi},
  booktitle = {Proceedings of the 8th International Sanskrit Computational Linguistics Symposium},
  year = {2026},
  url = {https://aclanthology.org/2026.iscls-1.3/}
}

References

26 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Hadi Abdine, Moussa Kamal Eddine, Davide Buscaldi, and Michalis Vazirgiannis. 2023. Word sense induction with agglomerative clustering and mutual information maximization. AI Open, 4:193–201. find
  2. Pushpak Bhattacharyya. 2010. IndoWordNet. In Nicoletta Calzolari, Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odijk, Stelios Piperidis, Mike Rosner, and Daniel Tapias, editors, Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC’10), Valletta, Malta… find
  3. Francis Bond and Kyonghee Paik. 2012. A survey of wordnets and their licenses. In proceedings of the 6th global WordNet conference (GWC 2012), pages 64–71. Matsue. find
  4. Pluempitiwiriyawej Charnyote, Cercone Nick, and An Xiangdong. 2009. Lexical acquisition and clustering of word senses to conceptual lexicon construction. Computers & Mathematics with Applications, 57(9):1537–1546. find
  5. Xinxiong Chen, Zhiyuan Liu, and Maosong Sun. 2015. Joint learning of character and word embeddings. In Proceedings of the 24th International Joint Conference on Artificial Intelligence (IJCAI). Relevant for gloss composition and context clustering methodologies. find
  6. Fangxiaoyu Feng, Yinfei Yang, Daniel Cer, Naveen Arivazhagan, and Wei Wang. 2022. Language-agnostic BERT sentence embedding. In Smaranda Muresan, Preslav Nakov, and Aline Villavicencio, editors, Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long P…
  7. Mario Giulianelli, Marco Del Tredici, and Raquel Fernández. 2024. To word senses and beyond: Inducing concepts with contextualized language models. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics, St. Julians, Malta, March. Association f… find
  8. Oliver Hellwig and Erica Biagetti. 2025. The sanskrit sembank. Language Resources and Evaluation, pages 1–24. find
  9. Oliver Hellwig, 2010–2025. The Digital Corpus of Sanskrit (DCS). find
  10. Masato Kikuchi, Masatsugu Ono, Toshioki Soga, Tetsu Tanabe, and Tadachika Ozono. 2024. Coarse-grained sense inventories based on semantic matching between english dictionaries. In 2024 11th International Conference on Advanced Informatics: Concept, Theory and Application (ICAICTA), pages 1–6.
  11. Aniruddha Kulkarni and Ted Pedersen. 2007. Leveraging dictionary definitions for word sense discrimination. In Proceedings of the 11th Conference on Computational Natural Language Learning (CoNLL-2007), pages 74–81, Prague, Czech Republic, June. Association for Computational Linguistics. find
  12. Diana McCarthy, Marianna Apidianaki, and Katrin Erk. 2016. Word sense clustering and clusterability. In Proceedings of the Fourth International Workshop on Metaphor in NLP, San Diego, California, June. Association for Computational Linguistics.
  13. George A Miller, Richard Beckwith, Christiane Fellbaum, Derek Gross, and Katherine J Miller. 1990. Introduction to wordnet: An on-line lexical database. International journal of lexicography, 3(4):235– 244. find
  14. Roberto Navigli. 2006. Meaningful clustering of senses helps boost word sense disambiguation performance. In Nicoletta Calzolari, Claire Cardie, and Pierre Isabelle, editors, Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for…
  15. Dhaval K Patel and Amba Kulkarni. 2024. Word sense alignment of Sanskrit lexica. In Arnab Bhattacharya, editor, Proceedings of the 7th International Sanskrit Computational Linguistics Symposium, pages 1–13, Auroville, Puducherry, India, February. Association for Computational Linguistics. find
  16. Anagha Pradeep and Radhika Mamidi. 2025. Sandarśana: A survey on sanskrit computational linguistics and digital infrastructure for sanskrit. ACM Comput. Surv., 57(10), May. find
  17. Amruta Purandare and Ted Pedersen. 2004. Word sense discrimination by clustering contexts in Low and High dimensional spaces. In Proceedings of the Eleventh Conference on European Chapter of the Association for Computational Linguistics (EACL), Barcelona, Spain, April. Association for Computational… find
  18. Qwen. 2025. Qwen3 technical report. find
  19. Nils Reimers and Iryna Gurevych. 2019. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, 11.
  20. J. Shao, P. Bakx, and K. Bontcheva. 2022. Word sense disambiguation using clustered sense labels. In Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), Marseille, France. European Language Resources Association. find
  21. Pushkar Singh, Sandeep Kumar Pandey, and Rajkiran Panuganti. 2024. Vyakyarth: A multilingual sentence embedding model for indic languages. GitHub. find
  22. Rion Snow, Brendan O’Connor, Daniel Jurafsky, and Andrew Y. Ng. 2010. Clustering dictionary definitions using Amazon Mechanical Turk. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 1138–1147, Cambridge, MA, October. Association for Computational Ling… find
  23. Kaitao Song, Xu Tan, Tao Qin, Jianfeng Lu, and Tie-Yan Liu. 2020. Mpnet: masked and permuted pre-training for language understanding. In Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20, Red Hook, NY, USA. Curran Associates Inc.
  24. Vincent A Traag, Ludo Waltman, and Nees Jan Van Eck. 2019. From louvain to leiden: guaranteeing well-connected communities. Scientific reports, 9(1):1–12. find
  25. Ludo Waltman and Nees Jan Van Eck. 2013. A smart local moving algorithm for large-scale modularity-based community detection. The European physical journal B, 86(11):471. find
  26. Yuan Zhang, Xia Wang, and Zhou Li. 2023. A method for constructing word sense embeddings based on word sense induction using Bi-LSTM and DPC. Applied Intelligence, 53. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/pradeep-2026-building-computational-sense-inventory-monier/