[目的/意义] 科技文献推荐系统作为连接科研人员与科技文献的重要桥梁，能为科研人员在增量式科技资源过载的樊笼里，提供有效的信息获取指引。为解决科技文献推荐中存在的文献间关联稀疏问题，可以引入蕴含科技背景知识的知识图谱，发掘科技文献间的高阶知识关联。[方法/过程] 研究借助Wikidata知识图谱和实体链接技术构建科技文献知识图谱，在此基础上，结合知识图谱嵌入、注意力机制、多层感知机技术，构建基于知识图谱的科技文献推荐模型，并基于CiteULike推荐数据集验证其有效性。[结果/结论] 基于AUC、MRR、nDGG@5三个评价指标的统计结果显示，研究所构建的推荐模型效果最优，且通过案例分析发现，模型可以有效发掘用户兴趣与候选文献间的高阶知识关联，提升推荐性能。
[Purpose/Significance] Scientific and technological literature recommendation system is an important bridge between scientific researchers and scientific and technological literature. It serves as an update for scientific researchers to provide effective information acquisition guidance for researchers facing incremental scientific and technological resource overload. In order to solve the problem of sparse connections between literature in the recommendation of scientific and technological literature, it is necessary to introduce background knowledge and bury the high-level knowledge correlation between literature. [Method/Process] This article comprehensively utilized Wikidata and entity linking technology to construct a knowledge graph. Based on this, it combined knowledge graph embedding, attention mechanism, and multi-layer perceptron to construct a scientific and technological literature recommendation model that integrated background knowledge. Its effectiveness was verified using the CiteULike dataset. [Result/Conclusion] The experimental results show that the constructed recommendation model achieved optimal results in AUC, MRR and nDCG@5 metrics. Through case analysis, it finds that background knowledge based on knowledge graphs can effectively explore high-order knowledge associations between user interest and candidate literature, improving recommendation performance.
[1] 张晓林. Library-Inside: AI赋能图书馆新质生产力的一种基础模型[J]. 中国图书馆学报, 2024, 50(3): 4-16. (ZHANG X L. Library-inside: a foundation model for AI-empowered new quality productive forces for library services[J]. Journal of library science in China, 2024, 50(3): 4-16.)
[2] MA X, WANG R. Personalized scientific paper recommendation based on heterogeneous graph representation[J]. IEEE access, 2019, 7: 79887-79894.
[3] 王淇, 刘忠麟, 向波. 面向科技文献的精准推荐引擎研究[J]. 中国电子科学研究院学报, 2022, 17(10): 983-990. (WANG Q, LIU Z L, XIANG B. Research on accurate recommendation engine for scientific literature[J]. Journal of China Academy of Electronics and Information Technology, 2022, 17(10): 983-990.)
[4] 徐勇, 司凤山, 吴延辉, 等. 基于概念泛化的科技文献推荐算法[J]. 图书情报工作, 2012, 56(21): 101-108. (XU Y, SI F S, WU Y H, et al. Concept generalization-based recommendation algorithm of academic papers[J]. Library and information service, 2012, 56(21): 101-108.)
[5] LOPEZ A I, SOSA S V J, ROJASJAS L F, et al. Improving selection of synsets from WordNet for domain-specific word sense disambiguation[J]. Computer speech & language, 2017, 41(C): 128-145．
[6] PRADHAN T, PAL S. A multi-level fusion based decision support system for academic collaborator recommendation[J]. Knowledge-based systems, 2020, 197: 105784.
[7] ZHU Y, LIN Q, LU H, et al. Recommending scientific paper via heterogeneous knowledge embedding based attentive recurrent neural networks[J]. Knowledge-based systems, 2021, 215: 106744.
[8] WINOTO P, TANG T Y, MCCALLA G. Contexts in a paper recommendation system with collaborative filtering[J]. International review of research in open and distributed learning, 2020, 13(5): 56-75.
[9] 李锴君, 牛振东, 时恺泽, 等. 基于学术知识图谱及主题特征嵌入的论文推荐方法[J]. 数据分析与知识发现, 2023, 7(5): 48-59. (LI K J, NIU Z D, SHI K Z, et al. Paper recommendation based on academic knowledge graph and subject feature embedding[J]. Data analysis and knowledge discovery, 2023, 7(5): 48-59.)
[10] MANJU G, ABHINAYA P, HAMALATHA M R, et al. Cold start problem alleviation in a research paper recommendation system using the random walk approach on a heterogeneous user-paper graph[J]. International journal of intelligent information technologies, 2020, 16(2): 24-48.
[11] LIU H, KOU H, YAN C, et al. Keywords-driven and popularity-aware paper recommendation based on undirected paper citation graph[J]. Complexity, 2020(1): 2085638.
[12] LI D, LIN C, GAO W, et al. Short-term IOT data forecast of urban public bicycle based on the DBSCAN-TCN model for social governance[J]. Wireless communications and mobile computing, 2021(1): 9488369.
[13] 李志峰. 基于外部知识融合推理的常识问答方法研究[D]. 苏州: 苏州大学, 2024. (LI Z F. Research on external knowledge integrated reasoning for commonsense question answering[D]. Suzhou: Soochow University, 2024.)
[14] 李源, 洛桑嘎登, 蒋卫丽. 融合外部知识和位置信息的中文命名实体识别[J]. 计算机工程与应用, 2024, 60(22): 162-171. (LI Y, LUOSANG G D, JIANG W L. Chinese named entity recognition based on external knowledge graph and position information[J]. Computer engineering and applications, 2024, 60(22): 162-171.)
[15] LI B, PI D. Network representation learning: a systematic literature review[J]. Neural computing and applications, 2020, 32(21): 16647-16679.
[16] JI G, HE S, XU L, et al. Knowledge graph embedding via dynamic mapping matrix[C]//Proceedings of the 53rd annual meeting of the Association for Computational Linguistics, Stroudsburg: Association for Computational Linguistics, 2015: 687-696.
[17] WANG X, HE X, CAO Y, et al. KGAT: knowledge graph attention network for recommendation[C]//Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining. New York: Association for Computing Machinery, 2019: 950-958.
[18] WANG H, CHEN B, LI W J. Collaborative topic regression with social regularization for tag recommendation[C]//Twenty-third international joint conference on artificial intelligence. New York: Association for Computing Machinery, 2013: 2719-2725.
[19] 新智元. 吴恩达的二八定律[EB/OL]. [2025-03-01]. https://www.51cto.com/article/657024.html. (XIN Z Y. Andrew Ng’s the 80/20 Rule[EB/OL]. [2025-03-01]. https://www.51cto.com/article/657024.html.)
[20] 孙金杨, 刘柏嵩, 任豪, 等. TAGAN: 一种融合细粒度语义特征的学术论文对抗推荐算法[J]. 电信科学, 2021, 37(8): 57-65. (SUN J Y, LIU B S, REN H, et al. TAGAN: an academic paper adversarial recommendation algorithm incorporating fine-grained semantic features[J]. Telecommunication science, 2021, 37(8): 57-65.)
[21] SARWAR B, KARYPIS G, KONSTAN J, et al. Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the 10th international conference on World Wide Web. New York: Association for Computing Machinery, 2001: 285-295.
