pith. sign in

arxiv: 2607.01605 · v1 · pith:7PFDUKE5new · submitted 2026-07-02 · 💻 cs.DB

pykci: A Compact Urban Knowledge Graph for Semantic and Spatial Queries using LLMs

Huynh Duc An Son Nguyen , Lukas Arzoumanidis , Youness Dehbi This is my paper

Pith reviewed 2026-07-03 03:37 UTC · model grok-4.3

classification 💻 cs.DB
keywords CityGMLknowledge graphNeo4jtext-to-CypherLLMsemantic 3D city modelsspatial queriesurban planning
0
0 comments X

The pith

pykci converts CityGML semantic 3D city models into a Neo4j knowledge graph that LLMs can query in natural language.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents pykci as a pipeline that ingests full CityGML 2.0 datasets, including all thematic modules and levels of detail, into a compact knowledge graph stored in Neo4j with R-tree spatial indexing. A complete round-trip path allows lossless export back to CityGML and export to OGC 3D Tiles for visualization. Natural-language questions are handled by feeding the graph schema to an LLM, which generates Cypher queries executed directly against the stored city data. Answers are therefore drawn from the exact dataset rather than model parameters, supporting both local open-weight models and commercial ones. The system is shown on Hamburg LoD2 open data for tasks such as locating roof surfaces suitable for greening.

Core claim

pykci transforms CityGML 2.0 datasets into a compact urban knowledge graph in Neo4j whose schema covers every thematic feature module across all levels of detail and carries an R-tree spatial index. An end-to-end Python pipeline performs ingestion, 3D Tiles export, and lossless round-trip to CityGML. Querying occurs through a model-agnostic text-to-Cypher layer that supplies only the graph schema as context so that an LLM translates natural-language questions into executable Cypher statements whose results remain grounded in the city graph.

What carries the argument

The text-to-Cypher mechanism that supplies the graph schema as context to an LLM so the model can emit executable Cypher queries against the urban knowledge graph.

If this is right

  • Urban planners can locate roof surfaces for greening by asking questions in plain English.
  • Citizens and GIS users can retrieve semantic and spatial city information without learning database schemas or Cypher.
  • Responses carry direct provenance from the stored graph rather than model memory, reducing hallucination risk.
  • Sensitive city data can stay on-premise when a local open-weight model is used.
  • Existing CityGML workflows remain compatible because of the lossless round-trip export.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same schema-driven translation approach could be tested on other 3D city or building information standards.
  • Adding incremental update paths from live sensor feeds would allow queries over changing city conditions.
  • Combining the graph with web-based 3D viewers could let non-experts explore results visually in the same session.

Load-bearing premise

An LLM given only the graph schema as context can generate correct and complete Cypher queries for complex semantic and spatial questions without errors or hallucinations.

What would settle it

A benchmark set of complex urban queries run through the text-to-Cypher layer whose generated Cypher statements are executed and checked against manually verified results from the original CityGML files.

Figures

Figures reproduced from arXiv: 2607.01605 by Huynh Duc An Son Nguyen, Lukas Arzoumanidis, Youness Dehbi.

Figure 1
Figure 1. Figure 1: Urban knowledge graph of the HafenCity district, Hamburg, automatically generated by [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of pykci as the digital knowledge base of an urban digital twin. A CityGML dataset is ingested into an urban knowledge graph in Neo4j (1) and queried in natural language (2): the LLM receives the graph schema as context and accesses the database either by calling predefined functions or by generating Cypher queries directly. The graph can be exported to OGC 3D Tiles for visualization (3) and back … view at source ↗
Figure 3
Figure 3. Figure 3: An example of CityGML encoding (left) and com [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The pykci core graph schema, comprising five node categories: context (gray), feature (dark green), thematic (orange), and geometry (green); interior or nested nodes are omitted for clarity due to space constraints. Feature nodes drawn with a double border (Building, CityObject) contain a 2D bounding box and are spatially indexed using an R-tree. Edge properties preserve information required for lossless r… view at source ↗
Figure 5
Figure 5. Figure 5: Content composition of a CityGML Building object. Properties are grouped by their defining class, with the class hierarchy shown at the bottom. Property types follow the Java bindings of the citygml4j parser. All substantial content can be mapped to graph using pykci’s own graph schema: mapped to a dedicated node or subgraph, stored as a node property, stored verbatim as an XML subtree (opaque to queries),… view at source ↗
Figure 6
Figure 6. Figure 6: Visualization of the FZK-Haus CityGML dataset in [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Visualization of the Railway Scene dataset in LoD3 [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Mean mapping time (left) and database size (right) [PITH_FULL_IMAGE:figures/full_fig_p008_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Graph compactness of pykci compared to 3DCityKG for the same Hamburg CityGML LoD2 dataset. We evaluate the performance of pykci, 3DCityKG, and 3DCi￾tyDB using 20 predefined natural-language questions, comprising 10 citizen-oriented queries (e.g., roof materials, tallest buildings) and 10 expert-oriented queries (e.g., spatial aggregates, solar scor￾ing). The evaluation is conducted on the same Hamburg City… view at source ↗
Figure 10
Figure 10. Figure 10: Visualization of enriched metal roof material at [PITH_FULL_IMAGE:figures/full_fig_p009_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Semantic enrichment for buildings with diverse [PITH_FULL_IMAGE:figures/full_fig_p010_11.png] view at source ↗
read the original abstract

CityGML, the OGC standard for modeling, storage, and exchange of semantic 3D city models, describes urban objects with detailed semantics, geometry, and topology. Yet this richness is difficult to query directly: CityGML's XML encoding is designed for exchange rather than analysis, and relational mappings expose it through schemas requiring expert knowledge. We present pykci (Python Knowledge Graph for Cities), an open-source system that transforms CityGML 2.0 datasets into a compact urban knowledge graph in Neo4j and makes it queryable in natural language. The graph schema covers all thematic feature modules of CityGML 2.0 across all levels of detail and is spatially indexed with an R-tree for efficient geometric retrieval. A complete end-to-end Python pipeline ingests CityGML datasets into the knowledge graph, exports them to OGC 3D Tiles for interactive visualization, and supports lossless round-trip export of all content back to CityGML. For querying, the graph is paired with a large language model through a model-agnostic text-to-Cypher mechanism: the graph schema is supplied as context, and the model translates natural-language questions into Cypher queries executed against the graph. We evaluate both a locally running open-weight model, which keeps sensitive city data on-premise, and a state-of-the-art commercial model for the most demanding spatial and semantic queries. Answers are grounded in exact city data rather than the model's parametric memory, reducing hallucination and providing auditable provenance for every response. We demonstrate the system on open-government CityGML LoD2 datasets from Hamburg, Germany, including complex semantic and spatial queries such as identifying roof surfaces suitable for greening. pykci enables urban planners, GIS practitioners, and citizens to interact with semantic 3D city models without expertise in query languages and database schemas.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The paper presents pykci, an open-source Python system that ingests CityGML 2.0 datasets into a compact Neo4j knowledge graph covering all thematic modules and LoDs with R-tree spatial indexing, provides a model-agnostic LLM text-to-Cypher interface that supplies only the graph schema as context for natural-language queries, includes pipelines for 3D Tiles visualization export and lossless round-trip to CityGML, and demonstrates the system on Hamburg LoD2 open-government data for complex semantic-spatial queries such as roof-greening suitability. The central claim is that this enables urban planners, GIS practitioners, and citizens to interact with semantic 3D city models without expertise in query languages or schemas, with answers grounded in the graph rather than model memory.

Significance. If the text-to-Cypher translation is shown to be reliable, the system would offer a practical, open-source bridge between detailed CityGML models and non-expert users while preserving data provenance and supporting on-premise deployment. The reproducible pipeline, full schema coverage, spatial indexing, and round-trip fidelity are concrete strengths that could be adopted by the urban data community.

major comments (1)
  1. [Abstract] Abstract (evaluation paragraph): the manuscript states that both open-weight and commercial models 'were evaluated on demanding queries' and that 'answers are grounded in exact city data rather than the model's parametric memory, reducing hallucination,' yet reports no quantitative metrics (accuracy, recall, success rate on complex semantic+spatial queries, hallucination rate, or baseline comparisons). This directly undermines the load-bearing claim that the no-expertise guarantee holds.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback and for recognizing the strengths of the reproducible pipeline, schema coverage, spatial indexing, and round-trip fidelity. We address the single major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract (evaluation paragraph): the manuscript states that both open-weight and commercial models 'were evaluated on demanding queries' and that 'answers are grounded in exact city data rather than the model's parametric memory, reducing hallucination,' yet reports no quantitative metrics (accuracy, recall, success rate on complex semantic+spatial queries, hallucination rate, or baseline comparisons). This directly undermines the load-bearing claim that the no-expertise guarantee holds.

    Authors: We agree that the abstract's phrasing implies a quantitative evaluation that is not present in the manuscript. The current version demonstrates the system on Hamburg LoD2 data through qualitative examples of complex semantic-spatial queries but does not report numerical metrics such as translation success rate, accuracy, or hallucination rate. We will revise the manuscript by (1) updating the abstract to accurately describe the evaluation as a demonstration rather than a quantified benchmark, and (2) adding a new evaluation section that reports quantitative metrics on a curated set of demanding queries (success rate, exact-match Cypher correctness, and provenance verification). This revision will be included in the next version. revision: yes

Circularity Check

0 steps flagged

No circularity: system description with no derivations or self-referential predictions

full rationale

The paper presents a software pipeline that ingests CityGML into a Neo4j graph schema, adds spatial indexing, pairs it with an LLM for text-to-Cypher translation using the schema as context, and demonstrates queries on Hamburg data. No equations, fitted parameters, predictions derived from inputs, or load-bearing self-citations appear. The text-to-Cypher mechanism is described as a practical interface rather than a derived result; its correctness is an empirical claim (unquantified in the provided text) but does not reduce to a definitional or self-referential loop. The contribution is therefore self-contained as an engineering artifact.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

The paper relies on existing standards and technologies without introducing new fitted parameters or invented physical entities; the central contribution is an engineering integration.

axioms (3)
  • domain assumption CityGML 2.0 standard provides accurate and complete semantic, geometric, and topological descriptions of urban objects across all LoDs
    The ingestion and schema assume the input data follows the standard without loss or inconsistency.
  • domain assumption LLMs can reliably translate natural-language questions into correct Cypher queries when the graph schema is provided as context
    This underpins the querying capability and is stated as model-agnostic.
  • standard math Neo4j with R-tree indexing supports efficient retrieval for the described urban graph sizes
    Standard database assumption for spatial queries.

pith-pipeline@v0.9.1-grok · 5885 in / 1551 out tokens · 53517 ms · 2026-07-03T03:37:14.532588+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 31 canonical work pages · 1 internal anchor

  1. [1]

    Jethro Akroyd, Sebastian Mosbach, Amit Bhave, and Markus Kraft. 2021. Uni- versal Digital Twin - A Dynamic Knowledge Graph.Data-Centric Engineering2 (2021), e14. doi:10.1017/dce.2021.10

    work page doi:10.1017/dce.2021.10 2021

  2. [2]

    Renzo Angles, Marcelo Arenas, Pablo Barceló, Aidan Hogan, Juan Reutter, and Domagoj Vrgoč. 2017. Foundations of Modern Query Languages for Graph Databases.ACM Comput. Surv.50, 5, Article 68 (Sept. 2017), 40 pages. doi:10. 1145/3104031

    2017

  3. [3]

    2025.Amtliches 3D-Gebäudemodell in der Ausprägung Level of Detail 2 (LoD2-DE)

    Arbeitsgemeinschaft der Vermessungsverwaltungen der Länder der Bundesre- publik Deutschland (AdV). 2025.Amtliches 3D-Gebäudemodell in der Ausprägung Level of Detail 2 (LoD2-DE). https://www.adv-online.de/AdV-Produkte/Weitere- Produkte/3D-Gebaeudemodelle-LoD/

    2025

  4. [4]

    Nguyen, Lara Johannsen, Filip Rothaut, Weilian Li, and Youness Dehbi

    Lukas Arzoumanidis, Son H. Nguyen, Lara Johannsen, Filip Rothaut, Weilian Li, and Youness Dehbi. 2025. Object Detection for the Enrichment of Semantic 3D City Models with Roofing Materials.ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information SciencesX-4/W6-2025 (2025), 9–16. doi:10.5194/ isprs-annals-X-4-W6-2025-9-2025

    2025

  5. [5]

    Michael K. Bergman. 2019. A Common Sense View of Knowledge Graphs. Adaptive Information, Adaptive Innovation, Adaptive Infrastructure Blog. http: //www.mkbergman.com/2244/a-common-sense-view-of-knowledge-graphs/

    2019

  6. [6]

    Piero Andrea Bonatti, Stefan Decker, Axel Polleres, and Valentina Presutti. 2018. Knowledge Graphs: New Directions for Knowledge Representation on the Se- mantic Web (Dagstuhl Seminar 18371).Dagstuhl Reports8, 9 (2018), 29–111. https://doi.org/10.4230/DagRep.8.9.29

    work page doi:10.4230/dagrep.8.9.29 2018

  7. [7]

    Car, Timo Homburg, Matthew Perry, Frans Knibbe, Simon J

    Nicholas J. Car, Timo Homburg, Matthew Perry, Frans Knibbe, Simon J. D. Cox, Joseph Abhayaratna, Mathias Bonduel, Paul J. Cripps, and Krzysztof Janowicz. 2024.OGC GeoSPARQL - A Geographic Query Language for RDF Data. Open Geospatial Consortium (OGC). http://www.opengis.net/doc/IS/geosparql/1.1 OGC 22-047r1, Version 1.1, International Standard

    2024

  8. [8]

    Arkadiusz Chadzynski, Nenad Krdzavac, Feroz Farazi, Mei Qi Lim, Shiying Li, Ayda Grisiute, Pieter Herthogs, Aurel von Richthofen, Stephen Cairns, and Markus Kraft. 2021. Semantic 3D City Database - An Enabler for a Dynamic Geospatial Knowledge Graph.Energy and AI6 (2021), 100106. doi:10.1016/j. egyai.2021.100106

    work page doi:10.1016/j 2021

  9. [9]

    Shengyuan Chen, Chuang Zhou, Zheng Yuan, Qinggang Zhang, Zeyang Cui, Hao Chen, Yilin Xiao, Jiannong Cao, and Xiao Huang. 2026. You Don’t Need Pre-Built Graphs for RAG: Retrieval Augmented Generation with Adaptive Reasoning Structures.Proceedings of the AAAI Conference on Artificial Intelligence40, 36 (March 2026), 30270–30278. doi:10.1609/aaai.v40i36.40278

    work page doi:10.1609/aaai.v40i36.40278 2026

  10. [10]

    Simon Cox, Paul Daisey, Ron Lake, Clemens Portele, and Arliss Whiteside. 2004. OpenGIS(R) Geography Markup Language (GML) Implementation Specification. Open Geospatial Consortium (OGC). doi:10.13140/2.1.2846.2401 OGC 03-105r1, Version 3.1.1, Recommendation Paper

    work page doi:10.13140/2.1.2846.2401 2004

  11. [11]

    2023.OGC 3D Tiles Specification

    Patrick Cozzi and Sean Lilley. 2023.OGC 3D Tiles Specification. Open Geospatial Consortium (OGC). https://www.opengis.net/doc/cs/3D-Tiles/1.1 22-025r4, Version 1.1, Community Standard

    2023

  12. [12]

    Linfang Ding, Guohui Xiao, Albulen Pano, Mattia Fumagalli, Dongsheng Chen, Yu Feng, Diego Calvanese, Hongchao Fan, and Liqiu Meng. 2025. Integrating 3D City Data through Knowledge Graphs.Geo-spatial Information Science28, 2 (2025), 780–799. arXiv:https://doi.org/10.1080/10095020.2024.2337360 doi:10. 1080/10095020.2024.2337360

    work page doi:10.1080/10095020.2024.2337360 2025

  13. [13]

    Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, Dasha Metropolitansky, Robert Osazuwa Ness, and Jonathan Larson. 2025. From Local to Global: A Graph RAG Approach to Query-Focused Summarization. arXiv:2404.16130 [cs.CL] https://arxiv.org/abs/2404.16130

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  14. [14]

    Lisa Ehrlinger and Wolfram Wöß. 2016. Towards a Definition of Knowledge Graphs. InJoint Proceedings of the Posters and Demos Track of the 12th Interna- tional Conference on Semantic Systems - SEMANTiCS2016 and the 1st International Workshop on Semantic Change & Evolving Semantics (SuCCESS’16) co-located with the 12th International Conference on Semantic S...

    2016

  15. [15]

    http://ceur-ws.org/Vol-1695/paper4.pdf

  16. [16]

    Jie Feng, Tianhui Liu, Yuwei Du, Siqi Guo, Yuming Lin, and Yong Li. 2025. CityGPT: Empowering Urban Spatial Cognition of Large Language Models. InProceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2(Toronto ON, Canada)(KDD ’25). Association for Computing Machinery, New York, NY, USA, 591–602. doi:10.1145/3711896.3736878

    work page doi:10.1145/3711896.3736878 2025

  17. [17]

    Kolbe, Claus Nagel, and Karl-Heinz Häfele

    Gerhard Gröger, Thomas H. Kolbe, Claus Nagel, and Karl-Heinz Häfele. 2012. OGC City Geography Markup Language (CityGML) Encoding Standard. Open Geospatial Consortium (OGC). https://portal.ogc.org/files/?artifact_id=47842 OGC 12-019, Version 2.0.0, International Standard

    2012

  18. [18]

    Olaf Hartig. 2017. Foundations of RDF* and SPARQL* : (An Alternative Approach to Statement-Level Metadata in RDF). InProceedings of the 11th Alberto Mendelzon International Workshop on Foundations of Data Management and the Web 2017 : (CEUR Workshop Proceedings, Vol. 1912). Juan Reutter, Divesh Srivastava, Article

    2017

  19. [19]

    http://ceur-ws.org/Vol-1912/paper12.pdf

    1912

  20. [20]

    Suchanek, Klaus Berberich, Edwin Lewis-Kelham, Gerard de Melo, and Gerhard Weikum

    Johannes Hoffart, Fabian M. Suchanek, Klaus Berberich, Edwin Lewis-Kelham, Gerard de Melo, and Gerhard Weikum. 2011. Yago2: Exploring And Querying World Knowledge In Time, Space, Context, And Many Languages. InProceedings Of The 20th International Conference On World Wide Web, WWW 2011, Hyderabad, India, March 28 - April 1, 2011 (Companion Volume), Sadago...

    work page doi:10.1145/1963192.1963296 2011

  21. [21]

    Rashid, Anisa Rula, Lukas Schmelzeisen, Juan F

    Aidan Hogan, Eva Blomqvist, Michael Cochez, Claudia d’Amato, Gerard de Melo, Claudio Gutiérrez, Sabrina Kirrane, José Emilio Labra Gayo, Roberto Navigli, Se- bastian Neumaier, Axel-Cyrille Ngonga Ngomo, Axel Polleres, Sabbir M. Rashid, Anisa Rula, Lukas Schmelzeisen, Juan F. Sequeda, Steffen Staab, and Antoine Zim- mermann. 2021.Knowledge Graphs. Number 2...

    work page doi:10.2200/s01125ed1v01y202109dsk022 2021

  22. [22]

    2015.CityGML Example Railway Scene

    Karl-Heinz Häfele and Claus Nagel. 2015.CityGML Example Railway Scene. Institute for Automation and Applied Computer Science (IAI), Karlsruhe Institute of Technology (KIT) and virtualcitysystems GmbH. https://github.com/3dcitydb/ importer-exporter/tree/master/resources/samples

    2015

  23. [23]

    2021.CityGML Example FZK-Haus

    Institute for Automation and Applied Computer Science (IAI), Karlsruhe In- stitute of Technology (KIT). 2021.CityGML Example FZK-Haus. https: //www.citygmlwiki.org/index.php?title=FZK_Haus

    2021

  24. [24]

    Elmehdi Kanna, Jannik Matijevic, Lukas Arzoumanidis, Huynh Duc An Son Nguyen, and Youness Dehbi. 2026. Semantic Enrichment of 3D City Models via Roof Material Classification for Urban Greening and Heat Island Mitigation. SSRN Electronic Journal(2026). doi:10.2139/ssrn.6127041

    work page doi:10.2139/ssrn.6127041 2026

  25. [25]

    Kanna and T

    K. Kanna and T. H. Kolbe. 2025. Advanced User Interaction with Urban Digital Twins using Large Language Models.ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information SciencesX-G-2025 (2025), 469–476. doi:10.5194/ isprs-annals-X-G-2025-469-2025

    2025

  26. [26]

    Kanna and T

    K. Kanna and T. H. Kolbe. 2025. Automatic Enrichment of Semantic 3D City Models using Large Language Models.ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information SciencesX-4/W6-2025 (2025), 105–112. doi:10.5194/isprs-annals-X-4-W6-2025-105-2025

    work page doi:10.5194/isprs-annals-x-4-w6-2025-105-2025 2025

  27. [27]

    Bernd Ketzler, Vasilis Naserentin, Fabio Latino, Christopher Zangelidis, Liane Thuvander, and Anders Logg. 2020. Digital Twins for Cities: A State of the Art Review.Built Environment (1978-)46, 4 (2020), 547–573. http://www.jstor.org/ stable/45299343

    work page arXiv 2020

  28. [28]

    Kolbe, Tatjana Kutzner, Carl Steven Smyth, Claus Nagel, Carsten Roensdorf, and Charles Heazel

    Thomas H. Kolbe, Tatjana Kutzner, Carl Steven Smyth, Claus Nagel, Carsten Roensdorf, and Charles Heazel. 2021.OGC City Geography Markup Language (CityGML) Part 1: Conceptual Model Standard. Open Geospatial Consortium (OGC). https://www.opengis.net/doc/IS/CityGML-1/3.0 20-010, Version 3.0.0, International Standard

    2021

  29. [29]

    2019.CityJSON: A Compact and Easy-to-use Encoding of the CityGML Data Model

    Hugo Ledoux, Ken Arroyo Ohori, Kavisha Kumar, Balázs Dukai, Anna Labetski, and Stelios Vitalis. 2019.CityJSON: A Compact and Easy-to-use Encoding of the CityGML Data Model. Vol. 4. Springer Science and Business Media LLC. doi:10.1186/s40965-019-0064-0

    work page doi:10.1186/s40965-019-0064-0 2019

  30. [30]

    Mendes, Sebastian Hellmann, Mohamed Morsey, Patrick van Kleef, Sören Auer, and Christian Bizer

    Jens Lehmann, Robert Isele, Max Jakob, Anja Jentzsch, Dimitris Kontokostas, Pablo N. Mendes, Sebastian Hellmann, Mohamed Morsey, Patrick van Kleef, Sören Auer, and Christian Bizer. 2015. DBpedia – A Large-scale, Multilingual Knowledge Base Extracted From Wikipedia.Semantic Web Journal6, 2 (2015), 167–195. doi:10.3233/SW-140134

    work page doi:10.3233/sw-140134 2015

  31. [31]

    2023.Challenges of Urban Digital Twins: A Systematic Review and a Delphi Expert Survey

    Binyu Lei, Patrick Janssen, Jantien Stoter, and Filip Biljecki. 2023.Challenges of Urban Digital Twins: A Systematic Review and a Delphi Expert Survey. Vol. 147. Elsevier BV, 104716. doi:10.1016/j.autcon.2022.104716

    work page doi:10.1016/j.autcon.2022.104716 2023

  32. [32]

    Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020. Retrieval-Augmented Gen- eration for Knowledge-Intensive NLP Tasks. InAdvances in Neural Information Processing Systems, H. Larochelle, M. Ranzato, R. Hadsell, M...

    2020

  33. [33]

    Liu and C

    S. Liu and C. Wang. 2025. KCitychatBot: A Knowledge Graph Based Chatbot System for Large-scale CityGML Dataset.The International Archives of the Photogrammetry, Remote Sensing and Spatial Information SciencesXLVIII-4/W15- 2025 (2025), 99–105. doi:10.5194/isprs-archives-XLVIII-4-W15-2025-99-2025

    work page doi:10.5194/isprs-archives-xlviii-4-w15-2025-99-2025 2025

  34. [34]

    Yu Liu, Jingtao Ding, Yanjie Fu, and Yong Li. 2023. UrbanKG: An Urban Knowl- edge Graph System.ACM Trans. Intell. Syst. Technol.14, 4, Article 60 (May 2023), 25 pages. doi:10.1145/3588577

    work page doi:10.1145/3588577 2023

  35. [35]

    2025.3D-Gebäudemodell LoD2-DE Hamburg

    Metaver Metadatenverbund, Landesbetrieb Geoinformation und Vermes- sung (LGV) Hamburg. 2025.3D-Gebäudemodell LoD2-DE Hamburg. https: //metaver.de/trefferanzeige?cmd=doShowDocument&docuuid=2C1F2EEC- CF9F-4D8B-ACAC-79D8C1334D5E

    2025

  36. [36]

    2024.Automatic Detection and Interpretation of Changes in Massive Semantic 3D City Models

    Huynh Duc An Son Nguyen. 2024.Automatic Detection and Interpretation of Changes in Massive Semantic 3D City Models. Ph. D. Dissertation. Technical University of Munich. https://mediatum.ub.tum.de/1748695

    work page arXiv 2024

  37. [37]

    Nguyen and Thomas H

    Son H. Nguyen and Thomas H. Kolbe. 2024. Identification and Interpretation of Change Patterns in Semantic 3D City Models. InRecent Advances in 3D Geoinformation Science, Thomas H. Kolbe, Andreas Donaubauer, and Christof Beil (Eds.). Springer Nature Switzerland, 479–496. doi:10.1007/978-3-031-43699- Nguyen et al. 4_30 18th 3D GeoInfo Conference 2023, Techn...

    work page doi:10.1007/978-3-031-43699- 2024

  38. [38]

    S. H. Nguyen, Z. Yao, and T. H. Kolbe. 2017.Spatio-Semantic Comparison of Large 3D City Models in CityGML Using a Graph Database. Vol. IV-4/W5. Copernicus GmbH, 99–106. doi:10.5194/isprs-annals-iv-4-w5-99-2017 12th International 3D GeoInfo Conference 2017, University of Melbourne, Melbourne, Australia

    work page doi:10.5194/isprs-annals-iv-4-w5-99-2017 2017

  39. [39]

    Yansong Ning and Hao Liu. 2024. UrbanKGent: A Unified Large Language Model Agent Framework for Urban Knowledge Graph Construction. InAdvances in Neural Information Processing Systems, A. Globerson, L. Mackey, D. Belgrave, A. Fan, U. Paquet, J. Tomczak, and C. Zhang (Eds.), Vol. 37. Curran Associates, Inc., 123127–123154. doi:10.52202/079017-3913

    work page doi:10.52202/079017-3913 2024

  40. [40]

    Shirui Pan, Linhao Luo, Yufei Wang, Chen Chen, Jiapu Wang, and Xindong Wu

  41. [41]

    IEEE Transactions on Knowledge and Data Engineering36, 7 (2024), 3580–3599

    Unifying Large Language Models and Knowledge Graphs: A Roadmap. IEEE Transactions on Knowledge and Data Engineering36, 7 (2024), 3580–3599. doi:10.1109/TKDE.2024.3352100

    work page doi:10.1109/tkde.2024.3352100 2024

  42. [42]

    Carlos Ramonell, Rolando Chacón, and Héctor Posada. 2023. Knowledge Graph- based Data Integration System for Digital Twins of Built Assets.Automation in Construction156 (2023), 105109. doi:10.1016/j.autcon.2023.105109

    work page doi:10.1016/j.autcon.2023.105109 2023

  43. [43]

    O’Reilly Media, Inc

    Ian Robinson, Jim Webber, and Emil Eifrem. 2015.Graph Databases: New Oppor- tunities for Connected Data. " O’Reilly Media, Inc. "

    2015

  44. [44]

    Nicole Schneider, Nandini Ramachandran, Kent O’Sullivan, and Hanan Samet

  45. [45]

    InProceed- ings of the 4th ACM SIGSPATIAL International Workshop on Searching and Mining Large Collections of Geospatial Data (GeoSearch ’25)

    DistRAG: Towards Distance-Based Spatial Reasoning in LLMs. InProceed- ings of the 4th ACM SIGSPATIAL International Workshop on Searching and Mining Large Collections of Geospatial Data (GeoSearch ’25). Association for Computing Machinery, New York, NY, USA, 31–34. doi:10.1145/3764920.3771694

    work page doi:10.1145/3764920.3771694

  46. [46]

    Craig Taverner and Andreas Berger. 2025. Neo4j Spatial. https://github.com/ neo4j-contrib/spatial Accessed: June 10, 2026

    2025

  47. [47]

    United Nations Department of Economic and Social Affairs (UN DESA). 2019. World Urbanization Prospects: The 2018 Revision. United Nations. https://www.un- ilibrary.org/content/books/9789210043144

    work page arXiv 2019

  48. [48]

    Denny Vrandečić and Markus Krötzsch. 2014. Wikidata: A Free Collaborative Knowledgebase.Commun. ACM57, 10 (2014), 78–85. https://doi.org/10.1145/ 2629489

    2014

  49. [49]

    Mohammad Saif Wajid, Hugo Terashima-Marin, Peyman Najafirad, Santiago Enrique Conant Pablos, and Mohd Anas Wajid. 2024. DTwin-TEC: An AI-based TEC District Digital Twin and Emulating Security Events by Leveraging Knowl- edge Graph.Journal of Open Innovation: Technology, Market, and Complexity10, 2 (2024), 100297. doi:10.1016/j.joitmc.2024.100297

    work page doi:10.1016/j.joitmc.2024.100297 2024

  50. [50]

    Zhu Wang, Fengxia Han, and Shengjie Zhao. 2024. A Survey on Knowledge Graph Related Research in Smart City Domain.ACM Trans. Knowl. Discov. Data 18, 9, Article 223 (Nov. 2024), 31 pages. doi:10.1145/3672615

    work page doi:10.1145/3672615 2024

  51. [51]

    Olaf Wysocki, Benedikt Schwab, and Bruno Willenborg. 2022. Awesome CityGML: The Ultimate List of Open Data Semantic 3D City Models. Zenodo. doi:10.5281/zenodo.5899096 Version 1.0

    work page doi:10.5281/zenodo.5899096 2022

  52. [52]

    Z. Yao, C. Nagel, M. Kendir, B. Willenborg, and T. H. Kolbe. 2025. The New 3D City Database 5.0 - Advancing 3D City Data Management based on CityGML 3.0.ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences X-4/W6-2025 (2025), 241–248. doi:10.5194/isprs-annals-X-4-W6-2025-241-2025

    work page doi:10.5194/isprs-annals-x-4-w6-2025-241-2025 2025

  53. [53]

    Zhihang Yao, Claus Nagel, Felix Kunde, György Hudra, Philipp Willkomm, An- dreas Donaubauer, Thomas Adolphi, and Thomas H. Kolbe. 2018.3DCityDB - A 3D Geodatabase Solution for the Management, Analysis, and Visualization of Semantic 3D City Models based on CityGML. Vol. 3. Springer Science and Business Media LLC, 1–26. doi:10.1186/s40965-018-0046-7 A Query...

    work page doi:10.1186/s40965-018-0046-7 2018