Design and implementation of an automated web-based Igbo text analyzer using natural language processing (NLP) tools

Presently in the world, the Igbo language is one of the less-resourced languages because there are not many developed and easy-to-find digital resources for it. Digital resources such as Igbo text corpora, Igbo electronic dictionaries, Igbo morphological analyzers, and Igbo thesauri, which can analyze Igbo text documents, are very limited. This work aims to design and develop an automated Igbo text analyzer using Natural Language Processing (NLP) tools. The development of this web-based Igbo text analyzer involves the analysis of the lexical and grammatical characteristics of the Igbo language that aided the identification of the basic principles governing word change (inflection) in the Igbo language. The object-oriented hypermedia design methodology (OOHDM) was applied to segment the work into stages of conceptual design, navigational design, abstract interface design, and implementation. The system was implemented using ReactJS for the frontend and the Python Flask framework for the backend. Furthermore, SQLite and SQLiteStudio were used as the database and database management tools for the system. The Natural Language Toolkit (NLTK) was used for text document analysis to enable users to observe the frequency and statistical analysis of the Igbo text document, as well as the Part of Speech (POS) tags associated with the words of the language. The development of this Igbo text analyzer (IgboNLP web application) is a great step towards achieving the objectives of Basic Language Resources Kits (BLARK) for the language.