Graph neural networks (GNNs) are a type of deep learning model specifically designed for handling graph-structured data. Graphs are mathematical representations that capture the relationships and interactions among entities, such as nodes and edges in a network. In this systematic literature review, we provide an overview of the methods and applications of GNNs, based on a comprehensive analysis. This paper provides a general design pipeline for building GNN models, which includes data preprocessing, graph representation construction and model architecture design. The strengths of GNNs in capturing complex graph structures and modeling graph data in diverse applications including structured scenarios such as social networks, recommendation systems, chemistry, and biology, to non-structured scenarios such as images and text are also highlighted in this systematic literature review paper. Additionally, the limitations of GNNs, including scalability, robustness, handling complex graph structures, convergence, and interpretability are also identified.