EbisaAsfaw/Oromo_grammar

A Python Package that Auto-Generates a Grammar-aware Dataset from a Raw Text # Package name oromo_grammar package # Project Title A Python package that automatically generates grammar-aware Oromo phrases from a given raw text. ## Description This is a novel python package that recieves an Oromo text file from a user; extract every possible root-verbs from the text; and stores them into a python list. It then trims all the verbs in the list to form their corresponding list of stems. Finally, the algorithm builds the corresponding phrases from the stems using an appropriate affixation and personal pronouns. The package can be customized and be replicated for any other languages according to the gramatical structure of the language. It helps relief machine learning and NLP practitioners,by automatically generating a clean and grammar-aware dataset,which was used to be done only manually. The generated Oromo phrases can indicate sentence elements like, number, gender and cases. The automatically generated Oromo phrases can be applied to different machine learning applications. It can form a clean phrase-based dataset that can help to train a Neural Machine Translation (NMT). It can also form a word and phrase-based dictionery, that will help in automatic word/sentence completion. The output of this tool is also helpful for Linguistics, in providing a clean and correct gramatical information of a given language. ## Getting Started ### Dependencies * This software package can run on any platform/OS(Windows, Linux, or Mac) and no prerequisites is given, except the Python Software Specifications. * This software requires Python 3.8 or latter versions, with some useful liberary packages, like NumPy, for a smooth excusion. ### Installing * Before installing "oromo_grammar" package first check that you have Python 3 software,installed on your device. * To install Python 3 software on your machine, first download and install the latest python version from "python.org". * Install numPy liberary wi …