Md. Kowsher

Stevens Institute of Technology, Noakhali Science and Technology University

Papers

Total Citations

H-Index

About

Md. Kowsher is a researcher advancing natural language processing (NLP) for low-resource languages, with a primary focus on Bangla. His key contributions center on developing foundational language resources and tools to bridge the gap between Bangla and more widely-studied languages. His most cited work, "BnVec: Towards the Development of Word Embedding for Bangla Language Processing" (6 citations), introduces a pioneering word embedding model that enables machines to capture semantic relationships in Bangla text, a critical step for downstream NLP tasks like translation and sentiment analysis. He further strengthened the field with "BanglaLM: Data Mining based Bangla Corpus for Language Model Research" (5 citations) and its companion paper (2 citations), which provide large-scale, curated corpora for training language models—addressing the chronic data scarcity that hampers Bangla NLP research. By combining data mining techniques with rigorous corpus construction, Kowsher has laid essential groundwork for future language model development. His work not only demonstrates technical innovation but also underscores the importance of linguistic inclusivity in AI, making him a notable contributor to the democratization of NLP for underrepresented languages.