NLPashto: NLP Toolkit for Low-resource Pashto Language

In recent years, natural language processing (NLP) has transformed numerous domains, becoming a vital area of research. However, the focus NLP studies predomi-nantly centered on major languages like English, inadvertently neglecting low-resource Pashto. Pashto, spoken by population over 50 million worldwide, remains largely unexplored in research, lacking off-the-shelf resources and tools even for fundamental text-processing tasks. To bridge this gap, study presents NLPashto, an open-source publicly accessible toolkit specifically designed The initial version NLPashto introduces four state-of-the-art models Spelling Correction, Word Segmentation, Part-of-Speech (POS) Tagging, Offensive Language Detection. also includes essential pre-trained static word embeddings, Word2Vec, fastText, GloVe. Furthermore, we have monolingual model Pashto from scratch, using Bidirectional Encoder Representations Transformers (BERT) architecture. For training evaluation all models, developed several benchmark datasets included them toolkit. Experimental results demonstrate that exhibit satisfactory perfor-mance their respective This can be significant milestone will hopefully support speed-up future research field NLP.