结合大语言模型和知识感知重构的图增强服务推荐方法
Graph Enhancement Service Recommendation Method Combining Large Language Model and Knowledge Perception Reconstruction
DOI: 10.12677/csa.2024.1411217, PDF,    国家自然科学基金支持
作者: 张永健, 王 涛：广东工业大学自动化学院，广东 广州；广东省信息物理融合系统实验室，广东 广州；陈可洲, 程良伦：广东省信息物理融合系统实验室，广东 广州；广东工业大学计算机学院，广东 广州
关键词: LLM；知识图谱；知识感知重构；服务推荐；LLM； Knowledge Graph； Knowledge Perception Reconstruction； Service Recommendation
摘要: 近年来，基于知识图谱(KG)的服务推荐方法取得了显著进展，但现有方法的性能仍受限于节点信息属性的质量和图结构的局限性，尤其是在实体描述信息不足和大量无关三元组噪声的影响下，推荐准确性受到影响。为此，我们提出了一种结合大语言模型(LLM)的知识结构更新服务推荐方法。基于思维链设计提示策略，进一步增强了大语言模型在扩展知识图数据描述信息中的输出效果，从而全面提升了实体数据的语义表达能力。此外，设计了一种知识感知重构任务，该任务有效识别知识图谱中的关键关联子图，并削弱无用三元组对推荐结果的负面影响，进而优化图结构以提升推荐性能。在真实数据集上进行实验结果表明，我们的方法在面向mashup的API服务推荐优于几种先进的基线方法。
Abstract: In recent years, service recommendation methods based on knowledge graph (KG) have made remarkable progress, but the performance of existing methods is still limited by the quality of node information attributes and the limitations of graph structure, especially under the influence of insufficient entity description information and a lot of irrelevant triplet noise, which affects the recommendation accuracy. Therefore, we propose a knowledge structure update service recommendation method combined with a large language model (LLM). The suggestion strategy based on thought chain design further enhances the output effect of large language model in expanding knowledge graph data description information, thus comprehensively improving the semantic expression ability of entity data. In addition, a knowledge perception reconstruction task is designed, which effectively identifies the key association subgraphs in the knowledge graph, weakens the negative impact of useless triples on the recommendation results, and then optimizes the graph structure to improve the recommendation performance. Experimental results on real data sets show that our approach is superior to several advanced baseline approaches for Mashup-oriented API service recommendations.
文章引用：张永健, 陈可洲, 王涛, 程良伦. 结合大语言模型和知识感知重构的图增强服务推荐方法[J]. 计算机科学与应用, 2024, 14(11): 70-82. https://doi.org/10.12677/csa.2024.1411217

参考文献

[1] Agarwal, N., Sikka, G. and Awasthi, L.K. (2022) A Systematic Literature Review on Web Service Clustering Approaches to Enhance Service Discovery, Selection and Recommendation. Computer Science Review, 45, Article ID: 100498. [Google Scholar] [CrossRef] 
[2] Bhopale, S.D., Sahu, A. and Pandyaji, K.K. (2023) Web Services Recommendation System Using Machine Learning Algorithms. 2023 4th International Conference for Emerging Technology (INCET), Belgaum, 26-28 May 2023, 1-7. [Google Scholar] [CrossRef] 
[3] 王雅琦, 婷俞, 田梁, 等. 基于兼容性感知的服务推荐方法研究[J]. 软件工程与应用, 2024, 13(4): 447-454.
[4] 程芳颖, 婷俞, 林帅伽, 等. 基于Transformer的服务推荐方法研究[J]. 计算机科学与应用, 2024, 14(7): 35-41.
[5] Kwapong, B., Anarfi, R. and Fletcher, K.K. (2020) A Knowledge Graph Approach to Mashup Tag Recommendation. 2020 IEEE International Conference on Services Computing (SCC), Beijing, 7-11 November 2020, 92-99. [Google Scholar] [CrossRef] 
[6] Zhu, Y., Wang, X., Chen, J., Qiao, S., Ou, Y., Yao, Y., et al. (2024) LLMs for Knowledge Graph Construction and Reasoning: Recent Capabilities and Future Opportunities. World Wide Web, 27, Article No. 58. [Google Scholar] [CrossRef] 
[7] Zeng, A., Liu, X., Du, Z., et al. (2023) GLM-130B: An Open Bilingual Pre-Trained Model. arXiv: 2210.02414.
http://arxiv.org/abs/2210.02414
[8] Xia, L., Huang, C., Huang, C., Lin, K., Yu, T. and Kao, B. (2023) Automated Self-Supervised Learning for Recommendation. Proceedings of the ACM Web Conference 2023, Austin, 30 April-4 May 2023, 992-1002. [Google Scholar] [CrossRef] 
[9] Wang, X., He, X., Cao, Y., Liu, M. and Chua, T. (2019) KGAT: Knowledge Graph Attention Network for Recommendation. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, 4-8 August 2019, 950-958. [Google Scholar] [CrossRef] 
[10] Hu, Z., Dong, Y., Wang, K. and Sun, Y. (2020) Heterogeneous Graph Transformer. Proceedings of The Web Conference 2020, Taipei, 20-24 April 2020, 2704-2710. [Google Scholar] [CrossRef] 
[11] Wei, J., Wang, X., Schuurmans, D., et al. (2023) Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. arXiv: 2201.11903.
http://arxiv.org/abs/2201.11903
[12] Devlin, J., Chang, M.W., Lee, K. and Toutanova, K. (2019) BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding. arXiv: 1810.04805.
http://arxiv.org/abs/1810.04805
[13] Zhang, Z., Liu, X., Zhang, Y., Su, Q., Sun, X. and He, B. (2020) Pretrain-KGE: Learning Knowledge Representation from Pretrained Language Models. Findings of the Association for Computational Linguistics: EMNLP 2020, November 2020, 259-266. [Google Scholar] [CrossRef] 
[14] Ye, J., Chen, X., Xu, N., et al. (2023) A Comprehensive Capability Analysis of GPT-3 and GPT-3.5 Series Models. arXiv: 2303.10420.
http://arxiv.org/abs/2303.10420
[15] Touvron, H., Martin, L., Stone, K., et al. (2023) Llama 2: Open Foundation and Fine-Tuned Chat Models. arXiv: 2307.09288.
http://arxiv.org/abs/2307.09288
[16] Pan, S., Luo, L. and Wang, Y. (2024) Unifying Large Language Models and Knowledge Graphs: A Roadmap. IEEE Transactions on Knowledge and Data Engineering, 36, 3580-3599.
[17] Mruthyunjaya, V., Pezeshkpour, P. and Hruschka, E. (2023) Rethinking Language Models as Symbolic Knowledge Graphs. arXiv: 2308.13676.
http://arxiv.org/abs/2308.13676
[18] Xiao, Y., Li, C. and Thürer, M. (2023) A Patent Recommendation Method Based on KG Representation Learning. Engineering Applications of Artificial Intelligence, 126, Article ID: 106722. [Google Scholar] [CrossRef] 
[19] 陆倩平, 王红斌. 基于知识图谱融入注意力机制商品推荐的可解释性方法[J]. 计算机科学与应用, 2022, 12(6): 1506-1517.
[20] Cao, Y., Wang, X., He, X., Hu, Z. and Chua, T. (2019) Unifying Knowledge Graph Learning and Recommendation: Towards a Better Understanding of User Preferences. The World Wide Web Conference, San Francisco, 13-17 May 2019, 151-161. [Google Scholar] [CrossRef] 
[21] Wang, X., He, Q. and Liang, J. (2023) Language Models as Knowledge Embeddings. arXiv: 2206.12617.
http://arxiv.org/abs/2206.12617
[22] Jiang, P., Agarwal, S., Jin, B., Wang, X., Sun, J. and Han, J. (2023) Text Augmented Open Knowledge Graph Completion via Pre-Trained Language Models. Findings of the Association for Computational Linguistics: ACL 2023, Toronto, July 2023, 11161-11180. [Google Scholar] [CrossRef] 
[23] 张雅歌, 胡雪若白, 洁黄, 等. 一种基于知识图谱与内容的推荐算法[J]. 计算机科学与应用, 2022, 12(3): 662-672.
[24] Xiao, B., Xie, X., Yang, C. and Wang, Y. (2022) RTN-GNNR: Fusing Review Text Features and Node Features for Graph Neural Network Recommendation. IEEE Access, 10, 114165-114177. [Google Scholar] [CrossRef] 
[25] Wang, X., Huang, T., Wang, D., et al. (2024) Learning Intents behind Interactions with Knowledge Graph for Recommendation. arXiv: 2102.07057.
http://arxiv.org/abs/2102.07057
[26] Yang, Y., Huang, C., Xia, L. and Li, C.L. (2024) Knowledge Graph Contrastive Learning for Recommendation. arXiv: 2205.00976.
https://arxiv.org/abs/2205.00976
[27] Liu, W., Zhou, P., Zhao, Z., et al. (2019) K-BERT: Enabling Language Representation with Knowledge Graph. arXiv: 1909.07606.
http://arxiv.org/abs/1909.07606
[28] Feichtenhofer, C., Fan, H., Li, Y. and He, K.M. (2022) Masked Autoencoders as Spatiotemporal Learners. arXiv: 2205.09113.
http://arxiv.org/abs/2205.09113
[29] Bai, J., Bai, S. and Chu, Y. (2023) Qwen Technical Report. arXiv: 2309.01660.
http://arxiv.org/abs/2309.1660