{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T22:55:11Z","timestamp":1772578511335,"version":"3.50.1"},"reference-count":30,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T00:00:00Z","timestamp":1770595200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T00:00:00Z","timestamp":1772496000000},"content-version":"vor","delay-in-days":22,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Address parsing seeks to map noisy, abbreviated free-text addresses into standardized hierarchical tuples for large-scale information systems. Existing approaches struggle with semantic and structural ambiguity, hallucination from unconstrained generation, and deployment constraints under privacy and governance requirements. We present AddrKG-LLM, a two-stage framework that combines knowledge-graph (KG)\u2013aware retrieval with schema-restricted large language model (LLM) decoding. First, contrastive learning over multi-view administrative graphs yields node embeddings that retrieve and re-rank a compact Top-K candidate set, thereby bounding the search space while preserving high gold coverage (Recall@K). Second, a candidate-restricted decoder running on-premises produces JSON-compliant outputs, enforcing single-candidate field consistency and alignment with KG priors to improve controllability and policy compliance. Using de-identified real-world records, we evaluate structural consistency via micro-level accuracy (\n                    <jats:inline-formula>\n                      <jats:tex-math>$$A_{micro}$$<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    ) and macro-level accuracy (\n                    <jats:inline-formula>\n                      <jats:tex-math>$$A_{macro}$$<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    ), and assess system properties with Recall@K and latency. Across strong string-matching, sequence-labeling, and generic LLM baselines, AddrKG-LLM delivers consistent gains in\n                    <jats:inline-formula>\n                      <jats:tex-math>$$A_{micro}$$<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    and\n                    <jats:inline-formula>\n                      <jats:tex-math>$$A_{macro}$$<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    with a favorable Recall@K. The proposed method consists of three components: (i) multi-view graph aggregation, (ii) a hierarchy-aware self-supervised contrastive objective that derives positives\/negatives from administrative relations to align textual and graph embeddings, and (iii) candidate-restricted decoding within the KG-derived Top-K set. Overall, coupling KG-aware retrieval with constrained on-prem LLM decoding yields an accurate, controllable, and deployable solution for noisy-address structuring across domains.\n                  <\/jats:p>","DOI":"10.1038\/s41598-026-39348-z","type":"journal-article","created":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T21:34:58Z","timestamp":1770672898000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Patient address parsing via KG-aware contrastive learning and constrained on-prem LLM inference"],"prefix":"10.1038","volume":"16","author":[{"given":"Jinzhe","family":"Li","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xin","family":"Pan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanchao","family":"Jia","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,2,9]]},"reference":[{"issue":"1","key":"39348_CR1","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1186\/s12913-022-08574-2","volume":"22","author":"J Wu","year":"2022","unstructured":"Wu, J., Qiao, J., Nicholas, S., Liu, Y. & Maitland, E. The challenge of healthcare big data to china\u2019s commercial health insurance industry: evaluation and recommendations. BMC health services research 22(1), 1189 (2022).","journal-title":"BMC health services research"},{"key":"39348_CR2","doi-asserted-by":"publisher","unstructured":"Xue, Y. & Li, C. Extracting chinese geographic data from baidu map api. The Stata Journal: Promoting communications on statistics and Stata 20, 805\u2013811 (2020) https:\/\/doi.org\/10.1177\/1536867X20976313","DOI":"10.1177\/1536867X20976313"},{"key":"39348_CR3","doi-asserted-by":"crossref","unstructured":"Zhang, M. et al. Non-standard address parsing in chinese based on integrated chtoponer model and dynamic finite state machine. Applied Sciences 13(17) (2023).","DOI":"10.3390\/app13179855"},{"issue":"2","key":"39348_CR4","doi-asserted-by":"publisher","first-page":"334","DOI":"10.1111\/tgis.12522","volume":"23","author":"S Comber","year":"2019","unstructured":"Comber, S. & Arribas-Bel, D. Machine learning innovations in address matching: A practical comparison of word2vec and crfs. Transactions in GIS 23(2), 334\u2013348 (2019).","journal-title":"Transactions in GIS"},{"issue":"1","key":"39348_CR5","doi-asserted-by":"publisher","first-page":"11","DOI":"10.3390\/ijgi11010011","volume":"11","author":"P Cruz","year":"2021","unstructured":"Cruz, P., Vanneschi, L., Painho, M. & Rita, P. Automatic identification of addresses: A systematic literature review. ISPRS International Journal of Geo-Information 11(1), 11 (2021).","journal-title":"ISPRS International Journal of Geo-Information"},{"key":"39348_CR6","doi-asserted-by":"crossref","unstructured":"Devlin, J., Chang, M.-W., Lee, K. & Toutanova, K. Bert: Pre-training of deep bidirectional transformers for language understanding. In: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (long and Short Papers), pp. 4171\u20134186 (2019).","DOI":"10.18653\/v1\/N19-1423"},{"key":"39348_CR7","unstructured":"Li, H., Lu, W., Xie, P. & Li, L. Neural chinese address parsing. In: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pp. 3421\u20133431 (2019)."},{"key":"39348_CR8","doi-asserted-by":"crossref","unstructured":"Li, P. et al. A multi-semantic feature fusion method for complex address matching of chinese addresses. ISPRS International Journal of Geo-Information 14(6) (2025).","DOI":"10.3390\/ijgi14060227"},{"key":"39348_CR9","doi-asserted-by":"crossref","unstructured":"Chen, Y. & Chen, X. Research on knowledge graph application technology. In: Journal of Physics: Conference Series, vol. 1187, p. 042083 (2019). IOP Publishing","DOI":"10.1088\/1742-6596\/1187\/4\/042083"},{"issue":"1","key":"39348_CR10","doi-asserted-by":"publisher","first-page":"0000417","DOI":"10.1371\/journal.pdig.0000417","volume":"3","author":"J Gallifant","year":"2024","unstructured":"Gallifant, J. et al. Peer review of gpt-4 technical report and systems card. PLOS digital health 3(1), 0000417 (2024).","journal-title":"PLOS digital health"},{"key":"39348_CR11","unstructured":"Touvron, H. et al. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)."},{"key":"39348_CR12","doi-asserted-by":"crossref","unstructured":"Zhang, L., Liu, Y., Luo, Y., Gao, F. & Gu, J. Qwen-ig: A qwen-based instruction generation model for llm fine-tuning. In: Proceedings of the 2024 13th International Conference on Computing and Pattern Recognition, pp. 295\u2013302 (2024).","DOI":"10.1145\/3704323.3704357"},{"issue":"8081","key":"39348_CR13","doi-asserted-by":"publisher","first-page":"633","DOI":"10.1038\/s41586-025-09422-z","volume":"645","author":"D Guo","year":"2025","unstructured":"Guo, D. et al. Deepseek-r1 incentivizes reasoning in llms through reinforcement learning. Nature 645(8081), 633\u2013638 (2025).","journal-title":"Nature"},{"key":"39348_CR14","doi-asserted-by":"publisher","unstructured":"Reimers, N. & Gurevych, I. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In: Inui, K., Jiang, J., Ng, V., Wan, X. (eds.) Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 3982\u20133992. Association for Computational Linguistics, Hong Kong, China (2019). https:\/\/doi.org\/10.18653\/v1\/D19-1410 . https:\/\/aclanthology.org\/D19-1410\/","DOI":"10.18653\/v1\/D19-1410"},{"issue":"9","key":"39348_CR15","doi-asserted-by":"publisher","first-page":"421","DOI":"10.3390\/info11090421","volume":"11","author":"J Wang","year":"2020","unstructured":"Wang, J. & Dong, Y. Measurement of text similarity: a survey. Information 11(9), 421 (2020).","journal-title":"Information"},{"key":"39348_CR16","unstructured":"Panchendrarajan, R. & Amaresan, A.. Bidirectional LSTM-CRF for named entity recognition. In: Politzer-Ahles, S., Hsu, Y.-Y., Huang, C.-R., Yao, Y. (eds.) Proceedings of the 32nd Pacific Asia Conference on Language, Information and Computation. Association for Computational Linguistics, Hong Kong (2018). https:\/\/aclanthology.org\/Y18-1061\/"},{"key":"39348_CR17","doi-asserted-by":"publisher","unstructured":"Ma, X. & Hovy, E. End-to-end sequence labeling via bi-directional LSTM-CNNs-CRF. In: Erk, K., Smith, N.A. (eds.) Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 1064\u20131074. Association for Computational Linguistics, Berlin, Germany (2016). https:\/\/doi.org\/10.18653\/v1\/P16-1101 . https:\/\/aclanthology.org\/P16-1101\/","DOI":"10.18653\/v1\/P16-1101"},{"key":"39348_CR18","unstructured":"Akbik, A., Blythe, D. & Vollgraf, R. Contextual string embeddings for sequence labeling. In: Proceedings of the 27th International Conference on Computational Linguistics, pp. 1638\u20131649 (2018)."},{"issue":"12","key":"39348_CR19","doi-asserted-by":"publisher","first-page":"745","DOI":"10.3390\/ijgi9120745","volume":"9","author":"H Zhang","year":"2020","unstructured":"Zhang, H. et al. Recognition method of new address elements in chinese address matching based on deep learning. ISPRS International Journal of Geo-Information 9(12), 745 (2020).","journal-title":"ISPRS International Journal of Geo-Information"},{"key":"39348_CR20","doi-asserted-by":"crossref","unstructured":"Cao, Y. et al. Geo-encoder: A chunk-argument bi-encoder framework for chinese geographic re-ranking. In: Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 1516\u20131530 (2024).","DOI":"10.18653\/v1\/2024.eacl-long.91"},{"issue":"5","key":"39348_CR21","first-page":"665","volume":"47","author":"H Zhang","year":"2022","unstructured":"Zhang, H., Du, Q., Chen, Z. & Zhang, C. A chinese address parsing method using roberta-bilstm-crf. Geomatics and Information Science of Wuhan University 47(5), 665\u2013672 (2022).","journal-title":"Geomatics and Information Science of Wuhan University"},{"issue":"3","key":"39348_CR22","doi-asserted-by":"publisher","first-page":"859","DOI":"10.1111\/tgis.13055","volume":"27","author":"C Liu","year":"2023","unstructured":"Liu, C., He, X., Su, S. & Kang, M. A graph-based method for chinese address matching. Transactions in GIS 27(3), 859\u2013876 (2023).","journal-title":"Transactions in GIS"},{"issue":"14","key":"39348_CR23","doi-asserted-by":"publisher","first-page":"6162","DOI":"10.3390\/app14146162","volume":"14","author":"P Feng","year":"2024","unstructured":"Feng, P., Zhang, B., Yang, L. & Feng, S. Entity-alignment interaction model based on chinese roberta. Applied sciences 14(14), 6162 (2024).","journal-title":"Applied sciences"},{"key":"39348_CR24","doi-asserted-by":"crossref","unstructured":"Liu, W., Zhou, P., Zhao, Z., Wang, Z., Ju, Q., Deng, H., Wang, P.: K-bert: Enabling language representation with knowledge graph. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp. 2901\u20132908 (2020)","DOI":"10.1609\/aaai.v34i03.5681"},{"issue":"2","key":"39348_CR25","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3703155","volume":"43","author":"L Huang","year":"2025","unstructured":"Huang, L. et al. A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions. ACM Transactions on Information Systems 43(2), 1\u201355 (2025).","journal-title":"ACM Transactions on Information Systems"},{"key":"39348_CR26","doi-asserted-by":"publisher","unstructured":"Wan, Z., Cheng, F., Mao, Z., Liu, Q., Song, H., Li, J., Kurohashi, S.: GPT-RE: In-context learning for relation extraction using large language models. In: Bouamor, H., Pino, J., Bali, K. (eds.) Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp. 3534\u20133547. Association for Computational Linguistics, Singapore (2023). https:\/\/doi.org\/10.18653\/v1\/2023.emnlp-main.214 . https:\/\/aclanthology.org\/2023.emnlp-main.214\/","DOI":"10.18653\/v1\/2023.emnlp-main.214"},{"issue":"6","key":"39348_CR27","doi-asserted-by":"publisher","first-page":"1091","DOI":"10.1109\/TPAMI.2007.1078","volume":"29","author":"L Yujian","year":"2007","unstructured":"Yujian, L. & Bo, L. A normalized levenshtein distance metric. IEEE transactions on pattern analysis and machine intelligence 29(6), 1091\u20131095 (2007).","journal-title":"IEEE transactions on pattern analysis and machine intelligence"},{"key":"39348_CR28","doi-asserted-by":"crossref","unstructured":"Wang, Y., Qin, J., Wang, W.: Efficient approximate entity matching using jaro-winkler distance. In: International Conference on Web Information Systems Engineering, pp. 231\u2013239 (2017). Springer","DOI":"10.1007\/978-3-319-68783-4_16"},{"issue":"1","key":"39348_CR29","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1631\/BF02842477","volume":"6","author":"Z Yun-tao","year":"2005","unstructured":"Yun-tao, Z., Ling, G. & Yong-cheng, W. An improved tf-idf approach for text classification. Journal of Zhejiang University-Science A 6(1), 49\u201355 (2005).","journal-title":"Journal of Zhejiang University-Science A"},{"key":"39348_CR30","unstructured":"Yang, A., Li, A., Yang, B., Zhang, B., Hui, B., Zheng, B., Yu, B., Gao, C., Huang, C., Lv, C., et al.: Qwen3 technical report. arXiv preprint arXiv:2505.09388 (2025)"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-39348-z","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-39348-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-39348-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T22:02:54Z","timestamp":1772575374000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-39348-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,9]]},"references-count":30,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["39348"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-39348-z","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,9]]},"assertion":[{"value":"1 December 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 February 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 February 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}},{"value":"Due to the retrospective nature of the study, the Medical Ethics Committee of Civil Aviation General Hospital, Beijing, China, waived the need of obtaining approval and informed consent (approval No. 2025-L-K-057). All methods were carried out in accordance with relevant guidelines and regulations.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval"}}],"article-number":"8003"}}