Compilation, Analysis and Application of a Comprehensive Bangla Corpus KUMono

Abstract

Research in Natural Language Processing (NLP) and computational linguistics highly depends on a good quality representative corpus of any specific language. Bangla is one of the most spoken languages in the world but Bangla NLP research is in its early stage of development due to the lack of quality public corpus. This article describes the detailed compilation methodology of a comprehensive monolingual Bangla corpus, KUMono. The newly developed corpus consists of more than 350 million word tokens and more than one million unique tokens from 18 major text categories of online Bangla websites. We have conducted several word-level and character-level linguistic phenomenon analyses based on empirical studies of the developed corpus. The corpus follows Zipf's curve and hapax legomena rule. The quality of the corpus is also assessed by analyzing and comparing the inherent sparseness of the corpus with existing Bangla corpora, by analyzing the distribution of function words of the corpus and vocabulary growth rate. We have developed a Bangla article categorization application based on the KUMono corpus and received compelling results by comparing to the state-of-the-art models.

Creating and validating the Fine-Grained Question Subjectivity Dataset (FQSD): A new benchmark for enhanced automatic subjective question answering systems

“…Lexical metrics, such as the Total Word Count and Unique Word Count, offer insights into the textual richness and the diversity of the vocabulary. These metrics are crucial for evaluating the potential of datasets to provide varied linguistic input necessary for training robust NLP models [ 39 – 41 ]. By examining the total count of words alongside the count of unique words, the assessment not only evaluates the volume of linguistic content available but also its variety, which is indicative of the potential complexity and nuance that NLP models must grapple with.…”