The detection of program vulnerabilities remains a challenging task in software security. The existing vulnerability detection methods rarely consider the multidimensional feature space complementarity of program graph structures, which easily overlooks contextual environment features and syntax structure features. This disadvantage leads to insufficient performance in capturing complex structural features, which hinders the improvement in detection accuracy. To address this issue, this paper introduces a novel vulnerability detection method, EnGS2F, which adopts the representation learning of an enhanced graph structure to improve the efficiency of capturing vulnerability information. On the dimension of the graph structure, a context relationship graph (CRG) is integrated on the basis of a program dependency graph (PDG) to enrich the global structural context representation. On the dimension of graph nodes, abstract syntax tree (AST) embedding and paragraph embedding are integrated to solve the problem of insufficient feature space complementarity. Moreover, the combination of a gated graph neural network (GGNN) with a graph attention mechanism further improves the learning performance of the enhanced graph structure. EnGS2F has been rigorously evaluated on program slices from open-source vulnerability datasets, demonstrating significant improvements over current competitive methods in detecting program vulnerabilities. Specifically, EnGS2F achieved a significant increase in the F1 score, outperforming existing technologies by 6%.

Software defect prediction models are of great importance in software testing, however, they also face the problem of model uninterpretability. Association rules have good accuracy and interpretability, being widely used in interpretable rule mining scenarios, but there are some common problems with current research: 1) Data unbalance seriously affects the accuracy of mined rules; 2) Most studies treat features as equally important and ignore feature contribution degree; 3) Classification by default rules easily reduces the accuracy of defect classification. Therefore, in the class unbalance scenario, we propose a weighted association rule based on the contribution degree of features, which solves the problem that defective rules are difficult to mine and considers the contribution degree of features. The process of rule generation, ranking, pruning and prediction is optimized according to the weighted support of the rules, and an ensemble model incorporating multiple rules is built. Experimental results on the PROMISE dataset show that the model proposed in this paper obtains an average F1 and MCC improvement of 6.4 % and 9.8 %, respectively, compared with current state-of-the-art classifiers; in terms of interpretability, rule-based interpretation in this paper can provide developers with better guidance on defect repair and risk avoidance compared with model-agnostic methods. From the experimental results, it can be concluded that the contribution degree of features helps to improve the quality of the rule set, and the construction of diversified rules can improve the accuracy of rule prediction.

Compared to conventional speech synthesis, end-to-end speech synthesis has achieved much better naturalness with more simplified system building pipeline. End-to-end framework can generate natural speech directly from characters for English. But for other languages like Chinese, recent studies have indicated that extra engineering features are still needed for model robustness and naturalness, e.g, word boundaries and prosody boundaries, which makes the front-end pipeline as complicated as the traditional approach. To maintain the naturalness of generated speech and discard language-specific expertise as much as possible, in Mandarin TTS, we introduce a novel self-attention based encoder with learnable Gaussian bias in Tacotron. We evaluate different systems with and without complex prosody information and results show that the proposed approach has the ability to generate stable and natural speech with minimum language-dependent front-end modules.