In the rapidly evolving field of artificial intelligence, the extraction of symbolic knowledge from neural networks has become a pivotal area of study. In this chapter, we delve into various methodologies that aim to decipher and translate the complex decision-making processes of neural networks into interpretable, human-understandable forms. It begins with an exploration of Direct Rule Extraction Approaches, including techniques like TREPAN, DeepRED, and REANN, which illuminate the hidden layers of neural processing. The chapter then transitions to Relational Knowledge Extraction Approaches, highlighting methods that uncover the intricate web of relationships and patterns within data. Through these discussions, the chapter sheds light on the crucial role of symbolic knowledge extraction in enhancing the transparency and accountability of AI systems, especially in critical applications such as healthcare and finance.