<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Bangla NLP | Farhan Noor Dehan</title><link>https://dehan001.github.io/tags/bangla-nlp/</link><atom:link href="https://dehan001.github.io/tags/bangla-nlp/index.xml" rel="self" type="application/rss+xml"/><description>Bangla NLP</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Sun, 01 Dec 2024 00:00:00 +0000</lastBuildDate><image><url>https://dehan001.github.io/media/icon_hu7729264130191091259.png</url><title>Bangla NLP</title><link>https://dehan001.github.io/tags/bangla-nlp/</link></image><item><title>Low-Resource Bangla NLP</title><link>https://dehan001.github.io/project/bangla-nlp/</link><pubDate>Sun, 01 Dec 2024 00:00:00 +0000</pubDate><guid>https://dehan001.github.io/project/bangla-nlp/</guid><description>&lt;p>This research compares graph-based models, transformers, and compact language models across Bangla sentiment, emotion, sarcasm, hate-speech, and fake-news classification tasks.&lt;/p></description></item><item><title>TinyLLM Efficacy in Low-Resource Language: An Experiment on Bangla Text Classification</title><link>https://dehan001.github.io/publication/tinyllm-bangla/</link><pubDate>Sun, 01 Dec 2024 00:00:00 +0000</pubDate><guid>https://dehan001.github.io/publication/tinyllm-bangla/</guid><description>&lt;p>This work evaluates large and compact language models on Bangla sentiment, sarcasm, emotion, hate-speech, and fake-news detection.&lt;/p></description></item><item><title>Investigating the Effectiveness of Graph-Based Algorithms for Bangla Text Classification</title><link>https://dehan001.github.io/publication/dehan-etal-2023-investigating/</link><pubDate>Thu, 07 Dec 2023 00:00:00 +0000</pubDate><guid>https://dehan001.github.io/publication/dehan-etal-2023-investigating/</guid><description>&lt;p>This study evaluates standalone and BERT-enhanced graph neural networks for Bangla sentiment, sarcasm, fake-news, hate-speech, and emotion classification.&lt;/p></description></item></channel></rss>