Title: Natural Language Processing in Modern Knowledge-Based Systems
Authors: Alaa K. AlDammagh and Samy S. Abu-Naser
Volume: 9
Issue: 8
Pages: 74-79
Publication Date: 2025/08/28
Abstract:
A knowledge-based system (KBS) is a computer program that solves complex problems by relying on a centralized information source. Traditional programs rely on procedural code, but a KBS expresses knowledge in a structured format, which is then evaluated by an inference engine to generate conclusions. The fundamental challenge for modern KBS is that while they rely on organized, machine-readable data, the vast majority of human information exists in an unstructured form, such as books, articles, and web pages. This paper argues that Natural Language Processing (NLP) is not only useful but also critically necessary to bridge this gap. NLP acts as a bridge, providing computational algorithms to automatically extract, interpret, and structure information from natural language. This review synthesizes various methods from recent literature for developing contemporary KBS that handle large volumes of data. It discusses key NLP techniques for knowledge acquisition and representation, such as Named Entity Recognition (NER), relation extraction, and event extraction, and their role in creating knowledge graphs. The review also highlights the strengths, limitations, and future directions of these methods, including the importance of addressing ethical gaps and resource inequality.