Sentence Level Sentiment Analysis for Nepali Language Using BERT. Shrestha, P., Shakya, A., Joshi, B., Satyal, S., & Panday, S. P. In Tripathi, A. K., Saha, A. K., & Shrivastava, V., editors, Proceedings of World Conference on Artificial Intelligence: Advances and Applications, volume 1909, of Lecture Notes in Networks and Systems (LNNS), pages 25–38, Cham, 2026. Springer Nature Switzerland. ![Sentence Level Sentiment Analysis for Nepali Language Using BERT [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex 1 download Sentiment analysis is a crucial task in natural language processing (NLP) that involves determining the sentiment expressed in a piece of text. Although extensive research has been conducted on sentiment analysis in widely spoken languages, there is a significant gap in resources and models for low-resource languages like Nepali. This study focuses on developing a sentiment analysis model tailored for the Nepali language using the Bidirectional Encoder Representations from Transformers (BERT) architecture, specifically the NepBERTa model and the Multilingual Bert model. The proposed approach involves fine-tuning a pre-trained BERT models on a Nepali sentiment dataset to classify text into positive, neutral, and negative sentiments. Various hyperparameters, such as learning rates and batch size, were systematically tuned to optimize the performance of the model. The evaluation metrics used include accuracy and the F1 score, ensuring a robust assessment of the model capacity to handle the nuances of the Nepali language.
@inproceedings{10.1007/978-3-032-22289-3_3,
abstract = {Sentiment analysis is a crucial task in natural language processing (NLP) that involves determining the sentiment expressed in a piece of text. Although extensive research has been conducted on sentiment analysis in widely spoken languages, there is a significant gap in resources and models for low-resource languages like Nepali. This study focuses on developing a sentiment analysis model tailored for the Nepali language using the Bidirectional Encoder Representations from Transformers (BERT) architecture, specifically the NepBERTa model and the Multilingual Bert model. The proposed approach involves fine-tuning a pre-trained BERT models on a Nepali sentiment dataset to classify text into positive, neutral, and negative sentiments. Various hyperparameters, such as learning rates and batch size, were systematically tuned to optimize the performance of the model. The evaluation metrics used include accuracy and the F1 score, ensuring a robust assessment of the model capacity to handle the nuances of the Nepali language.},
added-at = {2026-07-08T17:49:18.000+0200},
address = {Cham},
author = {Shrestha, Palisha and Shakya, Aman and Joshi, Basanta and Satyal, Sanjivan and Panday, Sanjeeb Prasad},
biburl = {https://www.bibsonomy.org/bibtex/2eacb10ba101f156776e4162a9f767a25/amanshakya},
booktitle = {Proceedings of World Conference on Artificial Intelligence: Advances and Applications},
doi = {https://doi.org/10.1007/978-3-032-22289-3_3},
editor = {Tripathi, Ashish Kumar and Saha, Apu Kumar and Shrivastava, Vivek},
interhash = {9b06a7c754c394110e425910485aa892},
intrahash = {eacb10ba101f156776e4162a9f767a25},
isbn = {978-3-032-22289-3},
keywords = {BERT Nepali myown sentiment},
pages = {25--38},
publisher = {Springer Nature Switzerland},
series = {Lecture Notes in Networks and Systems (LNNS)},
timestamp = {2026-07-08T17:49:18.000+0200},
title = {Sentence Level Sentiment Analysis for Nepali Language Using BERT},
url = {https://link.springer.com/chapter/10.1007/978-3-032-22289-3_3},
volume = 1909,
year = 2026
}
Downloads: 1
{"_id":"v3QFBzNMcubjXmYqb","bibbaseid":"shrestha-shakya-joshi-satyal-panday-sentencelevelsentimentanalysisfornepalilanguageusingbert-2026","author_short":["Shrestha, P.","Shakya, A.","Joshi, B.","Satyal, S.","Panday, S. P."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","abstract":"Sentiment analysis is a crucial task in natural language processing (NLP) that involves determining the sentiment expressed in a piece of text. Although extensive research has been conducted on sentiment analysis in widely spoken languages, there is a significant gap in resources and models for low-resource languages like Nepali. This study focuses on developing a sentiment analysis model tailored for the Nepali language using the Bidirectional Encoder Representations from Transformers (BERT) architecture, specifically the NepBERTa model and the Multilingual Bert model. The proposed approach involves fine-tuning a pre-trained BERT models on a Nepali sentiment dataset to classify text into positive, neutral, and negative sentiments. Various hyperparameters, such as learning rates and batch size, were systematically tuned to optimize the performance of the model. The evaluation metrics used include accuracy and the F1 score, ensuring a robust assessment of the model capacity to handle the nuances of the Nepali language.","added-at":"2026-07-08T17:49:18.000+0200","address":"Cham","author":[{"propositions":[],"lastnames":["Shrestha"],"firstnames":["Palisha"],"suffixes":[]},{"propositions":[],"lastnames":["Shakya"],"firstnames":["Aman"],"suffixes":[]},{"propositions":[],"lastnames":["Joshi"],"firstnames":["Basanta"],"suffixes":[]},{"propositions":[],"lastnames":["Satyal"],"firstnames":["Sanjivan"],"suffixes":[]},{"propositions":[],"lastnames":["Panday"],"firstnames":["Sanjeeb","Prasad"],"suffixes":[]}],"biburl":"https://www.bibsonomy.org/bibtex/2eacb10ba101f156776e4162a9f767a25/amanshakya","booktitle":"Proceedings of World Conference on Artificial Intelligence: Advances and Applications","doi":"https://doi.org/10.1007/978-3-032-22289-3_3","editor":[{"propositions":[],"lastnames":["Tripathi"],"firstnames":["Ashish","Kumar"],"suffixes":[]},{"propositions":[],"lastnames":["Saha"],"firstnames":["Apu","Kumar"],"suffixes":[]},{"propositions":[],"lastnames":["Shrivastava"],"firstnames":["Vivek"],"suffixes":[]}],"interhash":"9b06a7c754c394110e425910485aa892","intrahash":"eacb10ba101f156776e4162a9f767a25","isbn":"978-3-032-22289-3","keywords":"BERT Nepali myown sentiment","pages":"25–38","publisher":"Springer Nature Switzerland","series":"Lecture Notes in Networks and Systems (LNNS)","timestamp":"2026-07-08T17:49:18.000+0200","title":"Sentence Level Sentiment Analysis for Nepali Language Using BERT","url":"https://link.springer.com/chapter/10.1007/978-3-032-22289-3_3","volume":"1909","year":"2026","bibtex":"@inproceedings{10.1007/978-3-032-22289-3_3,\n abstract = {Sentiment analysis is a crucial task in natural language processing (NLP) that involves determining the sentiment expressed in a piece of text. Although extensive research has been conducted on sentiment analysis in widely spoken languages, there is a significant gap in resources and models for low-resource languages like Nepali. This study focuses on developing a sentiment analysis model tailored for the Nepali language using the Bidirectional Encoder Representations from Transformers (BERT) architecture, specifically the NepBERTa model and the Multilingual Bert model. The proposed approach involves fine-tuning a pre-trained BERT models on a Nepali sentiment dataset to classify text into positive, neutral, and negative sentiments. Various hyperparameters, such as learning rates and batch size, were systematically tuned to optimize the performance of the model. The evaluation metrics used include accuracy and the F1 score, ensuring a robust assessment of the model capacity to handle the nuances of the Nepali language.},\n added-at = {2026-07-08T17:49:18.000+0200},\n address = {Cham},\n author = {Shrestha, Palisha and Shakya, Aman and Joshi, Basanta and Satyal, Sanjivan and Panday, Sanjeeb Prasad},\n biburl = {https://www.bibsonomy.org/bibtex/2eacb10ba101f156776e4162a9f767a25/amanshakya},\n booktitle = {Proceedings of World Conference on Artificial Intelligence: Advances and Applications},\n doi = {https://doi.org/10.1007/978-3-032-22289-3_3},\n editor = {Tripathi, Ashish Kumar and Saha, Apu Kumar and Shrivastava, Vivek},\n interhash = {9b06a7c754c394110e425910485aa892},\n intrahash = {eacb10ba101f156776e4162a9f767a25},\n isbn = {978-3-032-22289-3},\n keywords = {BERT Nepali myown sentiment},\n pages = {25--38},\n publisher = {Springer Nature Switzerland},\n series = {Lecture Notes in Networks and Systems (LNNS)},\n timestamp = {2026-07-08T17:49:18.000+0200},\n title = {Sentence Level Sentiment Analysis for Nepali Language Using BERT},\n url = {https://link.springer.com/chapter/10.1007/978-3-032-22289-3_3},\n volume = 1909,\n year = 2026\n}\n\n","author_short":["Shrestha, P.","Shakya, A.","Joshi, B.","Satyal, S.","Panday, S. P."],"editor_short":["Tripathi, A. K.","Saha, A. K.","Shrivastava, V."],"key":"10.1007/978-3-032-22289-3_3","id":"10.1007/978-3-032-22289-3_3","bibbaseid":"shrestha-shakya-joshi-satyal-panday-sentencelevelsentimentanalysisfornepalilanguageusingbert-2026","role":"author","urls":{"Paper":"https://link.springer.com/chapter/10.1007/978-3-032-22289-3_3"},"keyword":["BERT Nepali myown sentiment"],"metadata":{"authorlinks":{}},"downloads":1},"bibtype":"inproceedings","biburl":"https://bibbase.org/f/EjJTJxsvFZCsHSath/amanshakya.bib","dataSources":["bhqr7NYGzPtbC6XAj"],"keywords":["bert nepali myown sentiment"],"search_terms":["sentence","level","sentiment","analysis","nepali","language","using","bert","shrestha","shakya","joshi","satyal","panday"],"title":"Sentence Level Sentiment Analysis for Nepali Language Using BERT","year":2026,"downloads":1}