Remembering software library components and mastering their application programming interfaces (APIs) is a daunting task for programmers, due to the sheer volume of available libraries. API completion tools, which predict subsequent APIs based on code context, are essential for improving development efficiency. Existing API completion techniques, however, face specific weaknesses that limit their performance. Pattern-based code completion methods that rely on statistical information excel in extracting common usage patterns of API sequences. However, they often struggle to capture the semantics of the surrounding code. In contrast, deep-learning-based approaches excel in understanding the semantics of the code but may miss certain common usages that can be easily identified by pattern-based methods. Our insight into overcoming these challenges is based on the complementarity between these two types of approaches. This paper proposes a combinatorial method of API completion that aims to exploit the strengths of both pattern-based and deep-learning-based approaches. The basic idea is to utilize a confidence-based selector to determine which type of approach should be utilized to generate predictions. Pattern-based approaches will only be applied if the frequency of a particular pattern exceeds a pre-defined threshold, while in other cases, deep learning models will be utilized to generate the API completion results. The results showed that our approach dramatically improved the accuracy and mean reciprocal rank (MRR) in large-scale experiments, highlighting its utility.

Code completion, a critical feature in integrated development environments, significantly reduces the coding workload for developers. Traditional code completion techniques often focus on the natural language properties of code, overlooking the structural characteristics of programming languages. To address this issue, this paper introduces a novel approach combining program analysis and deep learning to enhance the accuracy and efficiency of code completion. Initially, the research utilizes program analysis techniques to explore the structural and semantic information of code snippets deeply, constructing a program graph. Graph Gated Neural Networks (GGNN) and Transformer technologies are then employed to represent the program graph, capturing the local features and long-range dependencies of the code. The model integrates both the semantic and structural information of code, thereby providing more accurate completion suggestions. Experimental evaluations were conducted on public datasets for Python and JavaScript, two extensively used programming languages. The results demonstrate that our method significantly outperforms existing approaches in terms of code completion accuracy and Mean Reciprocal Rank (MRR).