Graph Neural Networks (GNNs) have been widely applied to software defect prediction with demonstrated promising performance. However, most existing approaches rely on simple graph representations of source code, making them inadequate for capturing high-order interaction patterns among multiple code entities and often overlook complex non-linear semantic and structural dependencies inherent in software systems. Furthermore, as software systems evolve dynamically across versions, inter-entity relationships undergo continuous changes, yet such cross-version evolutionary patterns are rarely considered in existing methods. To address these limitations, we propose a novel software defect prediction approach based on Temporal Hypergraph Neural Networks(THGNN2defect). We first parsed the source code into abstract syntax trees (ASTs) and extracted semantic features using convolution neural networks. Afterward, a semantic hypergraph based on semantic similarity among code entities and a structural hypergraph based on topological relationships in the class dependency graph are constructed, respectively. The extracted semantic and structural features are then combined to initialize node properties in the hypergraphs. Building upon this, semantic and structural hypergraphs from different versions are aligned and connected along a temporal axis to construct a temporal hypergraph that captures the dynamic evolution of code over time. Finally, we introduced a temporal hypergraph convolution and aggregation mechanism to integrate multi-version information, and derive the final semantic and structural representations of the code for defect prediction. We evaluated the proposed method on seven open-source projects by comparing with seven baseline approaches. The experimental results validate its effectiveness, showing that THGNN2defect achieves an average improvement of $\mathbf{2 \%} \boldsymbol{\sim} \mathbf{2 7. 1 \%}$ in F1-score and $\mathbf{1. 7 \% ~ 1 4. 2 \%}$ in AUC compared to the baselines.

Graph neural network is an effective deep learning framework for learning graph data.Existing research has introduced different variants of graph neural networks into the field of software defects and has achieved promising results.However, the graph neural network model based on the previous research is essentially transductive, is applied to a single fixed graph, and often ignores the direction and weight of the edges when modeling the network.In practice, software systems are dynamically evolving.Furthermore, in software network modeling, the direction and weight of edges are factors that are worth considering.Based on an inductive graph neural network, we proposed an improved defect prediction method named GSAGE2defect.We first constructed the class dependency network of the program and then used node2vec for embedding learning to automatically obtain the structural features of the network.Then we combined the learned structural features with traditional software code features to initialize the properties of nodes in the class dependency network.Next, we fed the dependency network to GraphSAGE for a deeper class representation.Finally, we evaluated the proposed method based on eight open-source programs and demonstrated that GSAGE2defect achieves an average improvement of 2.09%-26.69%over state-of-the-art methods in terms of F-measure.