Knowledge-Enhanced Program Repair for Data Science Code

This paper introduces DSrepair, a knowledge-enhanced program repair approach designed to repair the buggy code generated by LLMs in the data science domain. DSrepair uses knowledge graph based RAG for API knowledge retrieval and bug knowledge enrichment to construct repair prompts for LLMs. Specific...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings / International Conference on Software Engineering s. 898 - 910
Hlavní autoři: Ouyang, Shuyin, Zhang, Jie M., Sun, Zeyu, Penuela, Albert Merono
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 26.04.2025
Témata:
Code Repair
Codes
Computer bugs
Data science
Knowledge Graph
Knowledge graphs
Large Language Model
Large language models
Libraries
Maintenance engineering
Scalability
Software engineering
Syntactics
ISSN:1558-1225
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper introduces DSrepair, a knowledge-enhanced program repair approach designed to repair the buggy code generated by LLMs in the data science domain. DSrepair uses knowledge graph based RAG for API knowledge retrieval and bug knowledge enrichment to construct repair prompts for LLMs. Specifically, to enable knowledge graph-based API retrieval, we construct DS-KG (Data Science Knowledge Graph) for widely used data science libraries. For bug knowledge enrichment, we employ an abstract syntax tree (AST) to localize errors at the AST node level. We evaluate DSrepair's effectiveness against five state-of-the-art LLM-based repair baselines using four advanced LLMs on the DS-1000 dataset. The results show that DSrepair outperforms all five baselines. Specifically, when compared to the second-best baseline, DSrepair achieves substantial improvements, fixing 44.4%, 14.2%, 20.6%, and 32.1% more buggy code snippets for each of the four evaluated LLMs, respectively. Additionally, it achieves greater efficiency, reducing the number of tokens required per code task by 17.49%, 34.24%, 24.71%, and 17.59%, respectively.
ISSN:1558-1225
DOI:10.1109/ICSE55347.2025.00246