@inproceedings{rahman-etal-2022-cnn,
title = "{CNN} for Modeling {S}anskrit Originated {B}engali and {H}indi Language",
author = "Rahman, Chowdhury and
Rahman, MD. Hasibur and
Rafsan, Mohammad and
Ali, Mohammed Eunus and
Zakir, Samiha and
Muhammod, Rafsanjani",
editor = "He, Yulan and
Ji, Heng and
Li, Sujian and
Liu, Yang and
Chang, Chua-Hui",
booktitle = "Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = nov,
year = "2022",
address = "Online only",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.aacl-main.4/",
doi = "10.18653/v1/2022.aacl-main.4",
pages = "47--56",
abstract = "Though recent works have focused on modeling high resource languages, the area is still unexplored for low resource languages like Bengali and Hindi. We propose an end to end trainable memory efficient CNN architecture named CoCNN to handle specific characteristics such as high inflection, morphological richness, flexible word order and phonetical spelling errors of Bengali and Hindi. In particular, we introduce two learnable convolutional sub-models at word and at sentence level that are end to end trainable. We show that state-of-the-art (SOTA) Transformer models including pretrained BERT do not necessarily yield the best performance for Bengali and Hindi. CoCNN outperforms pretrained BERT with 16X less parameters and achieves much better performance than SOTA LSTMs on multiple real-world datasets. This is the first study on the effectiveness of different architectures from Convolution, Recurrent, and Transformer neural net paradigm for modeling Bengali and Hindi."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rahman-etal-2022-cnn">
<titleInfo>
<title>CNN for Modeling Sanskrit Originated Bengali and Hindi Language</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chowdhury</namePart>
<namePart type="family">Rahman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">MD.</namePart>
<namePart type="given">Hasibur</namePart>
<namePart type="family">Rahman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="family">Rafsan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammed</namePart>
<namePart type="given">Eunus</namePart>
<namePart type="family">Ali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samiha</namePart>
<namePart type="family">Zakir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rafsanjani</namePart>
<namePart type="family">Muhammod</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yulan</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Heng</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sujian</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chua-Hui</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online only</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Though recent works have focused on modeling high resource languages, the area is still unexplored for low resource languages like Bengali and Hindi. We propose an end to end trainable memory efficient CNN architecture named CoCNN to handle specific characteristics such as high inflection, morphological richness, flexible word order and phonetical spelling errors of Bengali and Hindi. In particular, we introduce two learnable convolutional sub-models at word and at sentence level that are end to end trainable. We show that state-of-the-art (SOTA) Transformer models including pretrained BERT do not necessarily yield the best performance for Bengali and Hindi. CoCNN outperforms pretrained BERT with 16X less parameters and achieves much better performance than SOTA LSTMs on multiple real-world datasets. This is the first study on the effectiveness of different architectures from Convolution, Recurrent, and Transformer neural net paradigm for modeling Bengali and Hindi.</abstract>
<identifier type="citekey">rahman-etal-2022-cnn</identifier>
<identifier type="doi">10.18653/v1/2022.aacl-main.4</identifier>
<location>
<url>https://aclanthology.org/2022.aacl-main.4/</url>
</location>
<part>
<date>2022-11</date>
<extent unit="page">
<start>47</start>
<end>56</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CNN for Modeling Sanskrit Originated Bengali and Hindi Language
%A Rahman, Chowdhury
%A Rahman, MD. Hasibur
%A Rafsan, Mohammad
%A Ali, Mohammed Eunus
%A Zakir, Samiha
%A Muhammod, Rafsanjani
%Y He, Yulan
%Y Ji, Heng
%Y Li, Sujian
%Y Liu, Yang
%Y Chang, Chua-Hui
%S Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
%D 2022
%8 November
%I Association for Computational Linguistics
%C Online only
%F rahman-etal-2022-cnn
%X Though recent works have focused on modeling high resource languages, the area is still unexplored for low resource languages like Bengali and Hindi. We propose an end to end trainable memory efficient CNN architecture named CoCNN to handle specific characteristics such as high inflection, morphological richness, flexible word order and phonetical spelling errors of Bengali and Hindi. In particular, we introduce two learnable convolutional sub-models at word and at sentence level that are end to end trainable. We show that state-of-the-art (SOTA) Transformer models including pretrained BERT do not necessarily yield the best performance for Bengali and Hindi. CoCNN outperforms pretrained BERT with 16X less parameters and achieves much better performance than SOTA LSTMs on multiple real-world datasets. This is the first study on the effectiveness of different architectures from Convolution, Recurrent, and Transformer neural net paradigm for modeling Bengali and Hindi.
%R 10.18653/v1/2022.aacl-main.4
%U https://aclanthology.org/2022.aacl-main.4/
%U https://doi.org/10.18653/v1/2022.aacl-main.4
%P 47-56
Markdown (Informal)
[CNN for Modeling Sanskrit Originated Bengali and Hindi Language](https://aclanthology.org/2022.aacl-main.4/) (Rahman et al., AACL-IJCNLP 2022)
ACL
- Chowdhury Rahman, MD. Hasibur Rahman, Mohammad Rafsan, Mohammed Eunus Ali, Samiha Zakir, and Rafsanjani Muhammod. 2022. CNN for Modeling Sanskrit Originated Bengali and Hindi Language. In Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 47–56, Online only. Association for Computational Linguistics.