Code search aims to retrieve semantically relevant code snippets on the basis of natural language queries. With the rapid expansion of public code repositories such as GitHub and Gitee, the efficient understanding and matching of relevant code have become critical challenges. Most existing deep learning-based code search approaches rely on feature extraction from code but often fail to account for its structural integrity. They neglect the complex hierarchical structure of the code, resulting in insufficient representational capacity for semantic matching. To address these challenges, we propose an innovative code search method, AcbertGraphC, which constructs a multi-relational graph representation by combining abstract syntax tree (AST), data dependency graph (DDG), and control flow graph (CFG) through a functional program graph and early fusion strategy. Additionally, we utilize meta-path aggregated graph neural networks (MAGNN) to extract complex relation-ships from the multi-relational graph, and leverage a graph attention mechanism to dynamically adjust meta-path selection, thereby enhancing the model’s search capability. The experi-mental results demonstrate that AcbertGraphC can accurately retrieve target code snippets and outperforms existing baseline methods in terms of matching precision.

Recent studies have demonstrated the potential of large language models (LLMs) in test case generation. However, LLMs often struggle to achieve high levels of test coverage when generating test cases for long methods. Long methods are one of the typical manifestations of code smells, characterized by excessive lines of code, complex control flows, deep nesting levels, and numerous variables. These characteristics make it difficult for traditional testing tools to cover most of the lines and branches of focal methods. To address this issue, this paper proposes a novel approach, DecoTest, to generate unit test cases for long methods based on LLMs. The proposed method leverages LLM and automated validation based on static analysis to derive high-quality decomposition and refactoring plans. After refactoring the long methods, test cases are generated, iteratively verified and repaired to produce the final test case suite. This paper also presents an experimental analysis of DecoTest. The results indicate that the proposed method outperforms existing LLM-based test case generation methods in terms of line coverage, branch coverage, and test execution pass rate.

Code search refers to the retrieval of code segments that best match the developer's needs from a large code base, whose needs are generally expressed in natural language. So far, the biggest difficulty in code search is the semantic gap between code language and natural language when matching, which is very different in grammar and language structure. In this paper, we propose a novel deep learning framework CodeFuse to achieve effective and accurate source code search by constructing a new mechanism to learn the rich semantics of source code and natural language queries from the two modalities of text and structure. We conduct extensive experiments using Java and Python corpus in CodeSearchNet which is large-scale and multi-language. Our multimodal approach surpasses baselines at most by 24.70% in MRR score, and 14.40% in NDCG score. The results show that our model can focus on information that is strongly correlated between code and query features without redundancy and improve the performance of code search.

Rule induction is an important approach for reasoning over Knowledge Graphs. Existing works mainly rely on searching for rule instances within the Knowledge Graph to induce rules. However, this approach may generate a vast search space, leading to inefficiency and difficulty in discovering rules that lack instance support. We propose a Logical Rule Induction based on Representation Learning (LRIRL) method, which can mine rules at the pattern level. By computing rule scores through vector representations of the rules, LRIRL can avoid the inefficiency caused by directly searching for rule instances in a vast search space. Furthermore, by incorporating the deductive nature of logical rules into the rule induction process, LRIRL can mine and evaluate rules even in the absence of rule instances. The generated rules can be used to perform more efficient and accurate reasoning tasks on the Knowledge Graph. Experimental results demonstrate that LRIRL outperforms baselines in both reasoning accuracy and rule mining efficiency on public datasets. Compared to the best baseline, LRIRL can achieve an average accuracy improvement of 3.43% in MRR, 1.40% in HITS@1, and 1.80% in HITS@10. Moreover, LRIRL can reduce rule mining time by an average of approximately 35% compared to the best baseline.

Knowledge Graph Reasoning based on Representation Learning maps the entities and relations into a vector space, assessing entity-pair similarities to deduce unknown facts. However, existing methods focus solely on the local importance of entities and ignore isolated entities, leading to the issue of missing feature information. To solve this, we propose Knowledge Representation Learning based on Global Information (KRLGI). KRLGI adopts an attention-based biased random walk algorithm to obtain global information and determine the importance of global entities. The global entity importance goes through a conversion into attention weights, and these weights are integrated with the local entity importance. Subsequently, the local and global entity importance can be used together to represent entity embeddings. Globally integrated representations can reveal richer semantics and enhance reasoning capabilities. Experimental results indicate that KRLGI outperforms other baselines in reasoning accuracy on four public datasets. Notably, on the FB15k-237 dataset, KRLGI shows significant improvements over the best baseline, with increases of 6.37% in MRR, 2.5% in HITS@1, and 5.5% in HITS@10.

Event causality identification is an important task in natural language processing. However, this task is highly challenging due to the high dependency of event context, text semantic ambiguity and insignificant causality features between text events. These issues lead to the low precision of causal relationship identification between events. We propose an Event Causality Identification Based on Feature Fusion (ECIFF) to improve the causality identification precision between events by integrating the context, semantics, and syntax of natural language. Firstly, we utilize BERT to capture the contextual features of events in natural language, enhancing the contextual embedding of events in different contexts. Secondly, based on an adversarial generative graph representation method, ECIFF learns a massive amount of causal relationships in the CauseNet, which can enhance the semantic representation of causes and effects of events. Next, we exploit the shortest dependency path to shorten the length of sentences and inductively learn all possible syntactic dependency relationships. Finally, the contextual, semantic and syntactic features are fused to synthetically determine the causal relationships among events. The experimental results indicate that our proposed approach significantly outperforms the state-of-the-art method LSIN: on the CTBank dataset, the precision, recall and F1-score of our approach are improved by 1.6%, 3.2% and 2.4%; on the ESL dataset, the precision, recall and F1-score of our approach are improved by 4.0%, 4.7% and 4.3%.

Causal relationships between events can reflect the historical evolution of events and provide an important reference for predicting future trends. Script prediction methods based on event graphs often struggle to adequately consider the complex interdependencies among events, leading to prediction biases. The Script Event Prediction Based on Causal Generalization Learning (SEPCG) method has been proposed to enhance the accuracy of script event prediction. SEPCG uses the graph attention network to learn the direct causal relationship similarity between known events and candidate events, the direct result event similarity between known events and candidate events, and utilizes double similarity generalization to learn the predicate type between known events and candidate events. SEPCG uses a Neural Tensor Network to learn parameter-level event embeddings and improve the model’s sensitivity to parameter-level changes. Finally, based on the generalized event embeddings, the BiLSTM network is used to simultaneously learn the forward contextual information from the known event to the candidate event direction, i.e., the cause to the result information, and the reverse contextual information from the candidate event to the known event direction, i.e., the result to the cause information. The BiLSTM is used to capture the temporal information of event chains at different levels. The effectiveness of the model is verified on the NYT dataset, with a 1.84% improvement in accuracy compared to the best baseline.