面向材料构效关系探索的领域专家指导知识挖掘框架

doi:10.15541/jim20260265

无机材料学报

面向材料构效关系探索的领域专家指导知识挖掘框架

左艳麟¹, 杨正伟², 盛浩强¹, 王达¹, 施思齐^1,3

1.上海大学核电关键材料全国重点实验室&材料科学与工程学院,上海 200444;
2.上海大学核电关键材料全国重点实验室&计算机工程与科学学院,上海 200444;
3.上海大学材料基因组工程研究院,上海 200444

收稿日期:2026-06-20 修回日期:2026-07-18
通讯作者: 杨正伟, 博士. E-mail: zhw_yann@shu.edu.cn; 施思齐, 教授. E-mail: sqshi@shu.edu.cn
作者简介:左艳麟(2001-), 女, 硕士生. E-mail: 23722267@shu.edu.cn

A Domain Expert-Guided Knowledge Mining Framework for Exploring Materials Structure-Property Relationships

ZUO Yanlin¹, YANG Zhengwei², SHENG Haoqiang¹, WANG Da¹, SHI Siqi^1,3

1. State Key Laboratory of Materials for Advanced Nuclear Energy & School of Materials Science and Engineering, Shanghai University, Shanghai 200444, China;
2. State Key Laboratory of Materials for Advanced Nuclear Energy & School of Computer Engineering and Science, Shanghai University, Shanghai 200444, China;
3. Materials Genome Institute, Shanghai University, Shanghai 200444, China

Received:2026-06-20 Revised:2026-07-18
Contact: YANG Zhengwei, PhD. E-mail: zhw_yann@shu.edu.cn; SHI Siqi, professor. E-mail: sqshi@shu.edu.cn
About author:ZUO Yanlin (2001-), female, Master candidate. E-mail: 23722267@shu.edu.cn
Supported by:
National Natural Science Foundation of China (92472207, 52472223, 92572301, 52372208)

摘要/Abstract

摘要： 知识挖掘能够从海量材料科学文献中自动提取潜在的科学规律,但其模型性能受限于标注数据的质量。然而,现有的标注范式通常难以准确描述关键的微观机制。为此,本研究提出了一种领域专家指导的知识挖掘框架,主要包含：用于标准化“成分-结构-性能”映射逻辑表达的本体约束规则;用于量化数据可学习性以把控标注质量的数据洞察评估策略;以及基于Neo4j的大规模三元组可视化与知识呈现图分析策略。本研究以具有复杂离子传输机制的复合固态电解质为例,展示了该框架在构建知识图谱中的应用。得益于融入领域知识的高质量数据集,训练后的知识挖掘模型实现了对837篇文献语料的自动化实体关系抽取,最终构建了包含两万余个节点的知识图谱。结果表明,填料的本征特性和多样的空间形貌引发了聚合物链段运动、连续导电通路形成以及界面力学增强,解释了复合固态电解质在离子电导率和力学性能上的提升机制。本工作不仅实现了从非结构化文本到高密度可视化信息的标准化转化,也为自动化知识挖掘以加速材料设计提供了一种可扩展的框架。本文数据集可在https://www.doi.org/10.57760/sciencedb.jim.00094中访问获取。

关键词: 知识挖掘, 知识图谱, 领域专家指导, 复合固态电解质, 成分-结构-性能关系

Abstract: Knowledge mining enables the automated extraction of latent scientific principles from large-scale materials science literature, of which performance is confined to the annotated data quality. However, current annotation paradigms usually fail to describe key microscopic mechanisms. Here, we propose a domain expert-guided knowledge mining framework, involving ontology-level constraints for standardizing composition-structure-property mapping logic representations, data insight evaluation strategy for quantifying data learnability to control annotation quality and graph analytics strategy based on Neo4j for visualizing large-scale triples and presenting knowledge. We demonstrate its application to building the knowledge graph of composite solid electrolytes (CSEs) with complex ionic transport mechanisms. Benefiting from the domain knowledge-incorporated dataset, a trained knowledge mining model automatizes the entity relationship extraction from a curated corpus of 837 publications and then a knowledge graph with over 20,000 nodes is constructed. Results indicate that the enhancement of CSEs in ionic conductivity and mechanical properties is explained through polymer chain segment motion, continuous conduction pathways and interfacial mechanical reinforcement, arising from fillers’ intrinsic characteristics and diverse spatial morphologies. This work not only enables ‌standardized‌ transformation of knowledge from unstructured text into high-density visualized information, but provides a scalable framework to automatize knowledge mining for accelerating materials design. The datasets in this article are listed in Science Data Bank at https://www.doi.org/10.57760/sciencedb.jim.00094.

Key words: knowledge mining, knowledge graph, domain expert guidance, composite solid-state electrolyte, composition-structure-property relationships

中图分类号:

TP181
TB3

左艳麟, 杨正伟, 盛浩强, 王达, 施思齐. 面向材料构效关系探索的领域专家指导知识挖掘框架[J]. 无机材料学报, DOI: 10.15541/jim20260265.

ZUO Yanlin, YANG Zhengwei, SHENG Haoqiang, WANG Da, SHI Siqi. A Domain Expert-Guided Knowledge Mining Framework for Exploring Materials Structure-Property Relationships[J]. Journal of Inorganic Materials, DOI: 10.15541/jim20260265.

参考文献

[1] JIANG X, WANG W, TIAN S, et al. Applications of natural language processing and large language models in materials discovery. npj Computational Materials, 2025, 11(1): 79.
[2] BUTLER K T, DAVIES D W, CARTWRIGHT H, et al. Machine learning for molecular and materials science. Nature, 2018, 559(7715): 547.
[3] XUE D, BALACHANDRAN P V, HOGDEN J, et al. Accelerated search for materials with targeted properties by adaptive design. Nature Communications, 2016, 7(1): 11241.
[4] RICKMAN J M, LOOKMAN T, KALININ S V.Materials informatics: from the atomic-level to the continuum.Acta Materialia, 2019, 168: 473.
[5] LIN T, CHEN S, HARRIS S J, et al. Investigating explainable transfer learning for battery lifetime prediction under state transitions. eScience, 2024, 4(5): 100280.
[6] TSHITOYAN V, DAGDELEN J, WESTON L, et al. Unsupervised word embeddings capture latent knowledge from materials science literature. Nature, 2019, 571(7763): 95-98.
[7] WANG W, JIANG X, TIAN S, et al. Automated pipeline for superalloy data by text mining. npj Computational Materials, 2022, 8(1): 9.
[8] LIU P, QIAN L, LU H, et al. The joint knowledge reasoning model based on knowledge representation learning for aviation assembly domain. Science China Technological Sciences, 2024, 67(1): 143.
[9] PEI Z, YIN J, ZHANG J.Language models for materials discovery and sustainability: progress, challenges, and opportunities.Progress in Materials Science, 2025, 154: 101495.
[10] SAMBASIVAN N, KAPANIA S, HIGHFILL H, et al.: “Everyone wants to do the model work, not the data work”: data cascades in high-stakes AI. Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, Yokohama, 2021: 1-15.
[11] HE B, WANG K, JIANG Z, et al. Towards a sustainable high-quality materials data ecosystem: frameworks and strategies. National Science Review, 2026, 13(7): nwag108.
[12] GRISHMAN R, SUNDHEIM B.Message understanding conference- 6: a brief history. The 16th International Conference on Computational Linguistics, Copenhagen, 1996: 466-471.
[13] AUGENSTEIN I, DAS M, RIEDEL S, et al. SemEval 2017 Task 10: ScienceIE - extracting keyphrases and relations from scientific publications. Proceedings of the 11th International Workshop on Semantic Evaluation, Vancouver, 2017: 546-555.
[14] LUAN Y, HE L, OSTENDORF M, et al.: Multi-task identification of entities, relations, and coreference for scientific knowledge graph construction. Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, 2018: 3219-3232.
[15] AU T W T, LAMPOS V, COX I. E-NER—An annotated named entity recognition corpus of legal text. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, Abu Dhabi, 2022: 246-255.
[16] FAESSLER E, MODERSOHN L, LOHR C, et al. ProGene - a large-scale, high-quality protein-gene annotated benchmark corpus. Proceedings of the 12th Language Resources and Evaluation Conference, Marseille, 2020: 4585-4596.
[17] TANABE L, XIE N, THOM L H, et al. GENETAG: a tagged corpus for gene/protein named entity recognition. BMC Bioinformatics, 2005, 6(1): S3.
[18] KIM J-D, OHTA T, TATEISI Y, et al. GENIA corpus—a semantically annotated corpus for bio-textmining. Bioinformatics, 2003, 19(Suppl 1): i180-182.
[19] CHOWDHURY M F M, LAVELLI A. Disease mention recognition with specific features. BioNLP '10: Proceedings of the 2010 Workshop on Biomedical Natural Language Processing, , Uppsala, 2010: 83-90.
[20] WEI C H, PENG Y, LEAMAN R, et al. Assessing the state of the art in biomedical relation extraction: overview of the BioCreative V chemical-disease relation (CDR) task. Database, 2016, 2016.
[21] ZHANG Y, XIAO G.Named entity recognition datasets: a classification framework.International Journal of Computational Intelligence Systems, 2024, 17(1): 71.
[22] EL-BOUSIYDY H, TRONCOSO J F, JOHANSSON P, et al. LIBAC: an annotated corpus for automated “reading” of the lithium-ion battery research literature. Chemistry of Materials, 2023, 35(5): 1849.
[23] FRIEDRICH A, ADEL H, TOMAZIC F, et al. The SOFC-exp corpus and neural approaches to information extraction in the materials science domain. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, online, 2020: 1255-1268.
[24] KONONOVA O, HUO H, HE T, et al. Text-mined dataset of inorganic materials synthesis recipes. Scientific Data, 2019, 6(1): 203.
[25] LIU Y, LIU D, GE X, et al. A high-quality dataset construction method for text mining in materials science. Acta Physica Sinica, 2023, 72(7): 070701.
[26] YANG X, LIU J, PEI N, et al. The critical role of fillers in composite polymer electrolytes for lithium battery. Nano-Micro Letters, 2023, 15(1): 74.
[27] ZHANG Z, CUI K, ZHANG J M, et al. Cellulose-based solid-state electrolytes for solid-state lithium metal batteries: advances in functionalization strategies and interfacial engineering. Advanced Energy Materials, 2026, 16(9): e04386.
[28] LU J, LIU Q, ZHENG Y, et al. Chain-segment ferry engineering from anchoring anion of the composite solid electrolyte enables fast lithium ion transport. Chemical Engineering Science, 2025, 303: 120962.
[29] Lü H, CHEN X, SUN Q, et al. Uniform garnet nanoparticle dispersion in composite polymer electrolytes. Acta Physico-Chimica Sinica, 2024, 40(3): 2305016.
[30] GUO Y, WU S, HE Y-B, et al. Solid-state lithium batteries: safety and prospects. eScience, 2022, 2(2): 138.
[31] WANG S, LA MONACA A, DEMOPOULOS G P.Composite solid-state electrolytes for all solid-state lithium batteries: progress, challenges and outlook.Energy Advances, 2024, 4(1): 11.
[32] LI J, PEI X, HE G, et al. Enhancing Li-ion transport by creating continuous channels and improving the decomposition of lithium salts in composite polymer electrolytes. Journal of Materiomics, 2025, 11(3): 100897.
[33] LIU Y, GE X, YANG Z, et al. An automatic descriptors recognizer customized for materials science literature. Journal of Power Sources, 2022, 545: 231946.
[34] LIU Y, LIU D, YANG Z, et al. A knowledge acquisition automatizing framework from literature exemplified by Na+ activation energy prediction of NASICON solid-state electrolyte. Energy Storage Materials, 2025, 80: 104390.
[35] LIU Y, YANG Z, ZOU X, et al. A general framework to govern machine learning oriented materials data quality. Materials Science and Engineering: R: Reports, 2025, 166: 101050.
[36] JI S, PAN S, CAMBRIA E, et al. A survey on knowledge graphs: representation, acquisition, and applications. IEEE Transactions on Neural Networks and Learning Systems, 2022, 33(2): 494.
[37] PAGE L, BRIN S, MOTWANI R, et al. The PageRank Citation Ranking: Bringing Order to the Web. The Web Conference, 1999: 1508503.
[38] BRIN S, PAGE L.The anatomy of a large-scale hypertextual web search engine.Computer Networks and ISDN Systems, 1998, 30(1): 107.
[39] FREEMAN L C.Centrality in social networks conceptual clarification.Social Networks, 1978, 1(3): 215.
[40] BREDE M.Networks—An Introduction. Mark E. J. Newman. (2010, Oxford University Press.). Artificial Life, 2012, 18(2): 241.
[41] BLONDEL V D, GUILLAUME J L, LAMBIOTTE R, et al. Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment, 2008, 2008: P10008.
[42] SHI S, GAO J, LIU Y, et al. Multi-scale computation methods: their applications in lithium-ion battery research and development. Chinese Physics B, 2016, 25(1): 018212.
[43] HE B, JIANG Z, WANG K, et al. Multirole collaborative and co-constructive materials design ecosystem enabled by using control and data flows decoupled workflows. Journal of Materiomics, 2026, 12(2): 101177.
[44] HE B, CHI S, YE A, et al. High-throughput screening platform for solid electrolytes combining hierarchical ion-transport prediction algorithms. Scientific Data, 2020, 7(1): 151.
[45] YANG Z, WU J, LIU D, et al. MATL-Miner: a lightweight joint model for automated knowledge extraction from materials science literature. Science China Materials, 2026.(in revision)
[46] ZHONG Z, CHEN D.A frustratingly easy approach for entity and relation extraction. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, online, 2021: 50-61.
[47] EBERTS M, ULGES A.Span-based joint entity and relation extraction with transformer pre-training. European Conference on Artificial Intelligence, Santiago de Compostela, 2020.
[48] GUPTA T, ZAKI M, KRISHNAN N M A, et al. MatSciBERT: a materials domain language model for text mining and information extraction. npj Computational Materials, 2022, 8(1): 102.