Hyperbolic Function Embedding: Learning Hierarchical Representation for Functions of Source Code in Hyperbolic Space

@article{Lu2019HyperbolicFE,
  title={Hyperbolic Function Embedding: Learning Hierarchical Representation for Functions of Source Code in Hyperbolic Space},
  author={Mingming Lu and Yan Liu and Haifeng Li and Dingwu Tan and Xiaoxian He and Wenjie Bi and Wendbo Li},
  journal={Symmetry},
  year={2019},
  volume={11},
  pages={254}
}
Recently, source code mining has received increasing attention due to the rapid increase of open-sourced code repositories and the tremendous values implied in this large dataset, which can help us understand the organization of functions or classes in different software and analyze the impact of these organized patterns on the software behaviors. Hence, learning an effective representation model for the functions of source code, from a modern view, is a crucial problem. Considering the… CONTINUE READING

Similar Papers

Figures, Results, and Topics from this paper.

Key Quantitative Results

  • HFE achieves up to 7.6% performance improvement compared to the chosen state-of-the-art methods, namely, Node2vec and Struc2vec.None Keywords: hyperbolic space; function-call graph; function embedding representation; source code mining 1.
  • Our method achieves up to 7.6% performance improvement (especially in low-dimension situations) compared with the chosen state-of-the-art embedding methods, namely, Node2vec [17] and Struc2vec [18].
  • Experiments show that HFE achieves up to 7.6% performance improvement compared to the chosen state-of-the-art graph embedding methods.

Citations

Publications citing this paper.

References

Publications referenced by this paper.