Science

A multi-relational graph perspective on semantic similarity in program retrieval

An summary of the proposed technique, together with: 1) Multi-relational graph building; 2) Multi-relational graph embedding; 3) Semantic similarity calculation. Credit: Frontiers of Pc Science (2023). DOI: 10.1007/s11704-023-2678-8

Program retrieval stays a cornerstone of software program growth, essential for reinforcing productiveness all through the event lifecycle. Amidst numerous program retrieval fashions, many have ignored the disparities between pure language queries and code, leading to a outstanding semantic hole.

Furthermore, packages and queries carry wealthy structural and semantic data. But, prevailing approaches typically overlook the cohesion amongst totally different points of supply code and deal with queries as sequences, neglecting their inherent structural traits.

To unravel these issues, a analysis staff led by Yunwei Dong printed their research in Frontiers of Pc Science.

The staff proposed a framework that formulates program retrieval as a multi-relational graph similarity drawback. Moreover, a dual-level consideration is utilized to assign weights to nodes in multi-relational graphs by intra-relation and inter-relation stage consideration.

To start, the multi-relational graph building module focuses on representing packages and queries utilizing code property graphs (CPG) and summary which means representations (AMR). This strategic strategy facilitates a extra complete and nuanced portrayal of program and question semantics.

Then the dual-level consideration graph neural network is leveraged to study semantic data for AMR and CPG. Lastly, a semantic similarity calculation module is designed to calculate the similarity of query-program pairs. In contrast with the prevailing analysis outcomes, the proposed technique performs comparatively nicely amongst all baselines.

Future analysis endeavors may think about optimizing multi-relational graphs by minimizing extraneous data, thereby diminishing graph complexity. Moreover, a promising avenue lies within the deliberate integration of exterior data, akin to data graphs, aiming to reinforce the illustration of program semantics.

Extra data:
Qianwen Gou et al, Semantic similarity-based program retrieval: a multi-relational graph perspective, Frontiers of Pc Science (2023). DOI: 10.1007/s11704-023-2678-8

Supplied by
Frontiers Journals

Quotation:
A multi-relational graph perspective on semantic similarity in program retrieval (2024, May 29)
retrieved 29 May 2024
from https://techxplore.com/information/2024-05-multi-graph-perspective-semantic-similarity.html

This doc is topic to copyright. Aside from any truthful dealing for the aim of personal examine or analysis, no
half could also be reproduced with out the written permission. The content material is offered for data functions solely.



Click Here To Join Our Telegram Channel


Source link

When you have any issues or complaints concerning this text, please tell us and the article will probably be eliminated quickly. 

Raise A Concern

Show More

Related Articles

Back to top button