pith. machine review for the scientific record. sign in

arxiv: 2604.22480 · v1 · submitted 2026-04-24 · 💻 cs.HC

Recognition: unknown

AI-based experts' knowledge visualization of cultural heritage: A case study of Terracotta Warriors

Siyi Li , Yue Jiang , Bowen Jing , Liuyuxin Yang , Yuhe Zhang
Authors on Pith no claims yet

Pith reviewed 2026-05-08 10:36 UTC · model grok-4.3

classification 💻 cs.HC
keywords cultural heritageTerracotta WarriorsAI visualizationdataset constructiongenerative adversarial networksrandom forestattribute analysisunified collection view
0
0 comments X

The pith

AI analysis of a new Terracotta Warriors dataset shows the collection as one entity with shared attribute patterns.

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

The paper builds a dataset that records attributes useful for distinguishing Terracotta Warriors from Pit No.1. It then uses generative adversarial networks to expand the data and random forests to detect distributions and relationships among those attributes, followed by visualizations that present the results. The central move is treating the warriors as a single group whose features can be seen together rather than as separate objects. A sympathetic reader would care because this unified view makes large heritage collections easier to grasp and compare, something single-figure displays cannot achieve.

Core claim

By constructing a dataset of attributes significant for identifying different Terracotta Warriors and applying generative adversarial networks and random forests to optimize the data, analyze distributions and relationships, and visualize the outcomes, the study presents the collection as a unified entity rather than isolated relics.

What carries the argument

An attribute dataset of Terracotta Warriors processed through a pipeline of generative adversarial networks for data augmentation and random forests for relationship analysis, followed by visualization of the results.

Load-bearing premise

The chosen attributes must be sufficient to capture meaningful differences and connections among the warriors, and the AI methods must accurately reveal real patterns in the data.

What would settle it

If the resulting visualizations reveal no distributions or relationships beyond what is already visible from descriptions of individual figures, the value of treating the collection as a unified entity would not be demonstrated.

Figures

Figures reproduced from arXiv: 2604.22480 by Bowen Jing, Liuyuxin Yang, Siyi Li, Yue Jiang, Yuhe Zhang.

Figure 1
Figure 1. Figure 1: Ranking of Feature Importance for the TW-1087S. accompanied by notable enhancements in precision and recall across most classes. Additionally, the AUC metric ascended from 85% to 98%, implying that employing data augmentation techniques on TW-1087 not only effectively addresses class imbalance concerns but also elevates the overall dataset quality. This enhancement holds significant importance in bolsterin… view at source ↗
Figure 2
Figure 2. Figure 2: Boxplots of the top six attributes in ranked importance in the TW-1087S. Box plots are a powerful tool for showing data dis￾tributions and statistics, providing a concise view of the central tendency and spread of the data across the range of a variable. The boxes represent attributes with multiple values, the lines correspond to individual values, and the small circles indicate outliers introduced during … view at source ↗
Figure 3
Figure 3. Figure 3: Violin plots of the six distinguishable attributes of TW-1087S. The x-axis delineates different classes of TW, while the y-axis illustrates the distribution of attribute values. Its width represents the data density within the numerical range; the wider it is, the denser the data. 5.3. Attribute Distribution Analysis and Visualization Although some of the attributes exhibit lower impor￾tance for classifica… view at source ↗
Figure 4
Figure 4. Figure 4: Attribute Correlation Matrix of the TW-1087S. measure of effect size for the chi-square test of indepen￾dence, used to assess the strength of the correlation between two categorical variables. Since the attribute "height" shows lower importance for classification and exhibits a less dis￾tinct distribution for each class, it is not included in this analysis. The correlation results of the attributes are vis… view at source ↗
Figure 5
Figure 5. Figure 5: The summary of the feedback of the questionnaire. we demonstrated how to interpret the figures by explaining all their elements, using examples that depicted the gender, major, and scores of students in a college to facilitate under￾standing. Participants were then given 8 minutes to explore the example figures independently. The interview lasted approximately 30 minutes, followed by the questionnaire proc… view at source ↗
read the original abstract

Advancements in 3D modeling,digital display technologies,and the growing availability of digital cultural heritage data have significantly improved the accuracy of heritage depictions and expanded opportunities for analysis.However,while many studies focus on presenting specific cultural heritage figurines,an often overlooked aspect is the visualization of the Terracotta Warriors as a unified entity.This involves concisely representing the distribution of features and their relationships,providing a clear and insightful presentation that engages practitioners, academics,and wider audiences.To tackle the challenges mentioned above,this research seeks to explore the application of AI methods in processing cultural heritage data.It aims to optimize and augment the dataset,analyze the distribution and relationships of various attributes, and interpret the analysis results through visualization techniques.The Terracotta Warriors,among China's most significant cultural heritages and renowned for their abundance,exquisite workmanship,and magnitude,are chosen as a case study.The contribution of this paper is primarily twofold.Firstly,we constructed a dataset of Terracotta Warriors from Pit No.1,detailing the attributes significant for identifying different Terracotta Warriors.Secondly,we employ various AI methods,such as generative adversarial network and random forest,to process and analyze these attributes,followed by visualizing the analysis results for an intuitive presentation.This study introduces a novel scheme for presenting information on a collection of cultural relics,offering a practical case for analyzing and visualizing the Terracotta Warriors'attributes as a whole entity,rather than showcasing individual relics'information in isolation.

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

2 major / 2 minor

Summary. The paper claims to introduce a novel visualization scheme for cultural heritage by constructing a dataset of attributes for Terracotta Warriors from Pit No.1 and applying AI methods (GAN for data augmentation/optimization and random forest for analysis of distributions and relationships) to present the collection as a unified entity rather than isolated relics, with results interpreted through visualizations for practitioners, academics, and audiences.

Significance. If the AI pipeline produces verifiable non-trivial insights into attribute distributions and relationships that standard statistical methods do not, and if the visualizations demonstrably improve understanding of the collection as a whole, the work could provide a practical template for digital cultural heritage analysis and presentation in HCI and digital humanities.

major comments (2)
  1. [Abstract] Abstract: The manuscript states that GAN and random forest are used to optimize/augment the dataset and analyze attribute distributions/relationships, yet supplies no concrete attribute list, no augmentation metrics (e.g., fidelity or diversity scores), no feature-importance or clustering outputs from the random forest, no example visualizations, and no comparison against non-AI baselines such as histograms or pairwise plots. This absence makes it impossible to evaluate whether the claimed 'whole entity' insights are achieved or are load-bearing for the central contribution.
  2. [Abstract] Abstract and contribution statement: The claim that the approach 'optimizes and augments the dataset' and 'reveals relationships' for an intuitive presentation rests on the unshown assumption that the constructed attributes are sufficient for identifying different warriors and that the chosen AI methods surface meaningful patterns; without reported results or validation, the novelty of the scheme over conventional collection-level visualization cannot be assessed.
minor comments (2)
  1. [Abstract] Abstract: Typographical errors include missing spaces (e.g., '3D modeling,digital display technologies,and', 'heritage depictions and expanded opportunities for analysis.However,while').
  2. [Abstract] Abstract: The phrase 'various AI methods, such as generative adversarial network and random forest' is vague; the paper should specify the full pipeline, justify the selection of these two techniques, and clarify what other methods (if any) were employed.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and constructive feedback on our manuscript. We address the major comments below and will make revisions to strengthen the presentation of our contributions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The manuscript states that GAN and random forest are used to optimize/augment the dataset and analyze attribute distributions/relationships, yet supplies no concrete attribute list, no augmentation metrics (e.g., fidelity or diversity scores), no feature-importance or clustering outputs from the random forest, no example visualizations, and no comparison against non-AI baselines such as histograms or pairwise plots. This absence makes it impossible to evaluate whether the claimed 'whole entity' insights are achieved or are load-bearing for the central contribution.

    Authors: We agree that the abstract lacks these specific details, which are important for evaluating the work. In the revised version, we will update the abstract to include a summary of the key attributes in the dataset (such as physical measurements, armor styles, and facial features), report quantitative metrics for the GAN augmentation (including fidelity and diversity scores), include feature importance rankings from the random forest analysis, provide example visualizations, and add a brief comparison to traditional statistical visualizations to highlight the benefits of the AI approach. revision: yes

  2. Referee: [Abstract] Abstract and contribution statement: The claim that the approach 'optimizes and augments the dataset' and 'reveals relationships' for an intuitive presentation rests on the unshown assumption that the constructed attributes are sufficient for identifying different warriors and that the chosen AI methods surface meaningful patterns; without reported results or validation, the novelty of the scheme over conventional collection-level visualization cannot be assessed.

    Authors: The manuscript does provide some description of the attributes in the methods section, but we acknowledge that the abstract and contribution statement do not sufficiently detail the validation or results. We will revise the abstract and introduction to explicitly state the attributes used and how they enable identification of warriors. Additionally, we will include validation results, such as accuracy of the random forest in identifying patterns and examples of revealed relationships, to demonstrate the novelty over conventional methods. revision: yes

Circularity Check

0 steps flagged

No significant circularity; descriptive case study with no derivations

full rationale

The paper constructs a dataset of Terracotta Warriors attributes and applies standard off-the-shelf AI methods (GAN for augmentation, random forest for analysis) before visualization. No equations, predictions, or first-principles claims appear in the provided text. The central contribution is an empirical workflow rather than any derived result that could reduce to its inputs by construction. No self-citations, fitted parameters renamed as predictions, or uniqueness theorems are invoked. This is a normal non-circular application paper.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are introduced; the work relies on standard machine learning algorithms and a custom dataset.

pith-pipeline@v0.9.0 · 5568 in / 1080 out tokens · 75643 ms · 2026-05-08T10:36:33.907403+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

48 extracted references · 28 canonical work pages

  1. [1]

    Bruno, S

    F. Bruno, S. Bruno, G. De Sensi, M.-L. Luchi, S. Mancuso, M. Muzzupappa, From 3d reconstruction to virtual reality: A complete methodology for digital archaeological exhibition, Journal of Cultural Heritage 11 (1) (2010) 42–49.doi:https: //doi.org/10.1016/j.culher.2009.02.006. URLhttps://www.sciencedirect.com/science/article/pii/ S1296207409000958

    work page doi:10.1016/j.culher.2009.02.006 2010

  2. [2]

    Scopigno, M

    R. Scopigno, M. Callieri, P. Cignoni, M. Corsini, M. Dellepiane, F. Ponchio, G. Ranzuglia, 3d models for cultural heritage: Beyond plain visualization., Computer 44 (7) (2011) 48–55

    2011

  3. [3]

    Zhang, W

    L. Zhang, W. Qi, K. Zhao, L. Wang, X. Tan, L. Jiao, Vr games and thedisseminationofculturalheritage,in:N.Streitz,S.Konomi(Eds.), Distributed, Ambient and Pervasive Interactions: Understanding Hu- mans, Springer International Publishing, Cham, 2018, pp. 439–451

    2018

  4. [4]

    Y. Ma, W. Li, N. An, Research on the digital exhibition of cultural relics for cross-culture communication: Case study on china-greece cooperation on terracotta warriors, ISPRS Annals of the Photogram- metry,RemoteSensingandSpatialInformationSciences(2023)187– 192

    2023

  5. [5]

    Díaz-Marín, E

    C. Díaz-Marín, E. Vendrell-Vidal, E. Aura-Castro, A. F. Abate, C. Sánchez-Belenguer, F. Narducci, Virtual reconstruction and rep- resentation of an archaeological terracotta statue, in: 2015 Digital Heritage, Vol. 2, 2015, pp. 699–702.doi:10.1109/DigitalHeritage. 2015.7419602

    work page doi:10.1109/digitalheritage 2015

  6. [6]

    Windhager, P

    F. Windhager, P. Federico, G. Schreder, K. Glinka, M. Dörk, S. Miksch, E. Mayr, Visualization of cultural heritage collection data: State of the art and future challenges, IEEE Transactions on VisualizationandComputerGraphics25(6)(2019)2311–2330.doi: 10.1109/TVCG.2018.2830759

    work page doi:10.1109/tvcg.2018.2830759 2019

  7. [7]

    Nickel, The first emperor and sculpture in china, Bulletin of the School of Oriental and African Studies 76 (3) (2013) 413–447

    L. Nickel, The first emperor and sculpture in china, Bulletin of the School of Oriental and African Studies 76 (3) (2013) 413–447

    2013

  8. [8]

    Towarek, L

    A. Towarek, L. Halicz, S. Matwin, B. Wagner, Machine learning in analytical chemistry for cultural heritage: A comprehensive review, Journal of Cultural Heritage 70 (2024) 64–70

    2024

  9. [9]

    3d digitizing of cultural heritage, Journal of Cultural Heritage 2 (1) (2001) 63–70.doi:https://doi.org/10.1016/S1296-2074(01)01108-6

    work page doi:10.1016/s1296-2074(01)01108-6 2001

  10. [10]

    Y. Hu, D. Lan, J. Wang, M. Hou, S. Li, X. Li, L. Zhu, Measurement and analysis of facial features of terracotta warriors based on high- precision 3d point clouds, Heritage Science 10 (1) (2022) 40

    2022

  11. [11]

    URLhttps://doi.org/10.1145/1658346.1658347 <Siyi Li et al.>:Preprint submitted to ElsevierPage 10 of 13 <TWVis>

    D.Koller,B.Frischer,G.Humphreys,Researchchallengesfordigital archives of 3d cultural heritage models 2 (3) (2010).doi:10.1145/ 1658346.1658347. URLhttps://doi.org/10.1145/1658346.1658347 <Siyi Li et al.>:Preprint submitted to ElsevierPage 10 of 13 <TWVis>

    work page doi:10.1145/1658346.1658347 2010

  12. [12]

    Pavlidis, A

    G. Pavlidis, A. Koutsoudis, F. Arnaoutoglou, V. Tsioukas, C. Chamzas, Methods for 3d digitization of cultural heritage, Journal of cultural heritage 8 (1) (2007) 93–98

    2007

  13. [13]

    Y. Wang, X. Deng, K. Zhang, Y. Lang, The intangible cultural heritage show mode based on ar technology in museums - take the li nationality non-material cultural heritage as an example, in: 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), 2018, pp. 936–940.doi:10.1109/ICIVC.2018.8492843

    work page doi:10.1109/icivc.2018.8492843 2018

  14. [14]

    An integrated vr/ar framework for user-centric interactive experience of cultural heritage: The arkaevision project, Digital Applications in Archaeology and Cultural Heritage 15 (2019) e00124.doi:https: //doi.org/10.1016/j.daach.2019.e00124

    work page doi:10.1016/j.daach.2019.e00124 2019

  15. [15]

    Ruecker, M

    S. Ruecker, M. Radzikowska, S. Sinclair, Visual interface design for digital cultural heritage: A guide to rich-prospect browsing, Ashgate Publishing, Ltd., 2011

    2011

  16. [16]

    Bitgood, Museum fatigue: A critical review, Visitor Studies 12 (2) (2009) 93–111.doi:10.1080/10645570903203406

    S. Bitgood, Museum fatigue: A critical review, Visitor Studies 12 (2) (2009) 93–111.doi:10.1080/10645570903203406

    work page doi:10.1080/10645570903203406 2009

  17. [17]

    A. R. Shaikh, D. Koop, H. Alhoori, M. Sun, Toward systematic design considerations of organizing multiple views, in: 2022 IEEE Visualization and Visual Analytics (VIS), 2022, pp. 105–109.doi: 10.1109/VIS54862.2022.00030

    work page doi:10.1109/vis54862.2022.00030 2022

  18. [18]

    Gleicher, D

    M. Gleicher, D. Albers, R. Walker, I. Jusufi, C. D. Hansen, J. C. Roberts, Visual comparison for information visualization, Information Visualization 10 (4) (2011) 289–309.doi:10.1177/ 1473871611416549

    2011

  19. [19]

    Javed, N

    W. Javed, N. Elmqvist, Exploring the design space of composite visualization,in:2012IEEEPacificVisualizationSymposium,2012, pp. 1–8.doi:10.1109/PacificVis.2012.6183556

    work page doi:10.1109/pacificvis.2012.6183556 2012

  20. [21]

    Franconeri

    X. Liu, H.-W. Shen, The effects of representation and juxtaposition ongraphicalperceptionofmatrixvisualization,AssociationforCom- putingMachinery,NewYork,NY,USA,2015.doi:10.1145/2702123. 2702217. URLhttps://doi.org/10.1145/2702123.2702217

    work page doi:10.1145/2702123 2015

  21. [23]

    Z. Qu, J. Hullman, Keeping multiple views consistent: Constraints, validations,andexceptionsinvisualizationauthoring,IEEETransac- tionsonVisualizationandComputerGraphics24(1)(2018)468–477. doi:10.1109/TVCG.2017.2744198

    work page doi:10.1109/tvcg.2017.2744198 2018

  22. [25]

    I. Cho, W. Dou, D. X. Wang, E. Sauda, W. Ribarsky, Vairoma: A visual analytics system for making sense of places, times, and events in roman history, IEEE Transactions on Visualization and Computer Graphics 22 (1) (2016) 210–219.doi:10.1109/TVCG.2015.2467971

    work page doi:10.1109/tvcg.2015.2467971 2016

  23. [26]

    Wagner, D

    M. Wagner, D. Slijepcevic, B. Horsak, A. Rind, M. Zeppelzauer, W.Aigner,Kavagait:Knowledge-assistedvisualanalyticsforclinical gait analysis, IEEE Transactions on Visualization and Computer Graphics 25 (3) (2019) 1528–1542.doi:10.1109/TVCG.2017.2785271

    work page doi:10.1109/tvcg.2017.2785271 2019

  24. [27]

    D. Han, G. Parsad, H. Kim, J. Shim, O.-S. Kwon, K. A. Son, J. Lee, I. Cho, S. Ko, Hisva: A visual analytics system for studying history, IEEE Transactions on Visualization and Computer Graphics 28 (12) (2022) 4344–4359.doi:10.1109/TVCG.2021.3086414

    work page doi:10.1109/tvcg.2021.3086414 2022

  25. [28]

    Kruskal,J.M

    J.B. Kruskal,J.M. Landwehr,Icicleplots: Betterdisplaysfor hierar- chical clustering, The American Statistician 37 (2) (1983) 162–168

    1983

  26. [29]

    Johnson, B

    B. Johnson, B. Shneiderman, Tree-maps: A space filling approach to the visualization of hierarchical information structures, Tech. rep. (1998)

    1998

  27. [30]

    57–65.doi:10.1109/INFVIS.2000.885091

    J.Stasko,E.Zhang,Focus+contextdisplayandnavigationtechniques for enhancing radial, space-filling hierarchy visualizations, in: IEEE SymposiumonInformationVisualization2000.INFOVIS2000.Pro- ceedings, 2000, pp. 57–65.doi:10.1109/INFVIS.2000.885091

    work page doi:10.1109/infvis.2000.885091 2000

  28. [31]

    Rodden, Applying a sunburst visualization to summarize user navigation sequences, IEEE Computer Graphics and Applications 34 (5) (2014) 36–40.doi:10.1109/MCG.2014.63

    K. Rodden, Applying a sunburst visualization to summarize user navigation sequences, IEEE Computer Graphics and Applications 34 (5) (2014) 36–40.doi:10.1109/MCG.2014.63

    work page doi:10.1109/mcg.2014.63 2014

  29. [32]

    832–836.doi:10.1109/BMEI.2015.7401618

    C.Liu,P.Wang,Asunburst-basedhierarchicalinformationvisualiza- tionmethodanditsapplicationinpublicopinionanalysis,in:20158th InternationalConferenceonBiomedicalEngineeringandInformatics (BMEI), 2015, pp. 832–836.doi:10.1109/BMEI.2015.7401618

    work page doi:10.1109/bmei.2015.7401618 2015

  30. [33]

    Médoc, H

    N. Médoc, H. Bourgoin, P. Pinheiro, M. Ghoniem, Using a multi- level and multi-resolution visual analytics software to understand the aftermath of a catastrophe, in: 2019 IEEE Conference on Visual Analytics Science and Technology (VAST), 2019, pp. 138–139.doi: 10.1109/VAST47406.2019.8986957

    work page doi:10.1109/vast47406.2019.8986957 2019

  31. [34]

    S. Liu, D. Maljovec, B. Wang, P.-T. Bremer, V. Pascucci, Visualizing high-dimensional data: Advances in the past decade, IEEE Transac- tions on Visualization and Computer Graphics 23 (3) (2017) 1249– 1268.doi:10.1109/TVCG.2016.2640960

    work page doi:10.1109/tvcg.2016.2640960 2017

  32. [35]

    L.Zhou,D.Weiskopf,Multivariatevisualizationofparticledata,The European Physical Journal Special Topics 227 (2019) 1741–1755

    2019

  33. [36]

    Nobre, N

    C. Nobre, N. Gehlenborg, H. Coon, A. Lex, Lineage: Visualizing multivariate clinical data in genealogy graphs, IEEE Transactions on Visualization and Computer Graphics 25 (3) (2019) 1543–1558. doi:10.1109/TVCG.2018.2811488

    work page doi:10.1109/tvcg.2018.2811488 2019

  34. [37]

    Inselberg, B

    A. Inselberg, B. Dimsdale, Parallel coordinates: a tool for visualiz- ing multi-dimensional geometry, in: Proceedings of the First IEEE Conference on Visualization: Visualization ‘90, 1990, pp. 361–378. doi:10.1109/VISUAL.1990.146402

    work page doi:10.1109/visual.1990.146402 1990

  35. [38]

    C. Viau, M. J. McGuffin, Y. Chiricota, I. Jurisica, The flowvizmenu and parallel scatterplot matrix: Hybrid multidimensional visualiza- tions for network exploration, IEEE Transactions on Visualization andComputerGraphics16(6)(2010)1100–1108.doi:10.1109/TVCG. 2010.205

    work page doi:10.1109/tvcg 2010

  36. [39]

    Tyagi, T

    A. Tyagi, T. Estro, G. Kuenning, E. Zadok, K. Mueller, Pc-expo: A metrics-based interactive axes reordering method for parallel coor- dinate displays, IEEE Transactions on Visualization and Computer Graphics 29 (1) (2023) 712–722.doi:10.1109/TVCG.2022.3209392

    work page doi:10.1109/tvcg.2022.3209392 2023

  37. [40]

    Fung, J.-K

    T.-L. Fung, J.-K. Chou, K.-L. Ma, A design study of personal biblio- graphic data visualization, in: 2016 IEEE Pacific Visualization Sym- posium (PacificVis), 2016, pp. 244–248.doi:10.1109/PACIFICVIS. 2016.7465279

    work page doi:10.1109/pacificvis 2016

  38. [41]

    A. Lex, M. Streit, E. Kruijff, D. Schmalstieg, Caleydo: Design and evaluation of a visual analysis framework for gene expression data in itsbiologicalcontext,in:2010IEEEPacificVisualizationSymposium (PacificVis), 2010, pp. 57–64.doi:10.1109/PACIFICVIS.2010.5429609

    work page doi:10.1109/pacificvis.2010.5429609 2010

  39. [42]

    Lhuillier, C

    A. Lhuillier, C. Hurter, A. Telea, Ffteb: Edge bundling of huge graphs by the fast fourier transform, in: 2017 IEEE Pacific Visual- ization Symposium (PacificVis), 2017, pp. 190–199.doi:10.1109/ PACIFICVIS.2017.8031594

    work page arXiv 2017

  40. [43]

    Q. W. Bouts, B. Speckmann, Clustered edge routing, in: 2015 IEEE PacificVisualizationSymposium(PacificVis),2015,pp.55–62.doi: 10.1109/PACIFICVIS.2015.7156356

    work page doi:10.1109/pacificvis.2015.7156356 2015

  41. [44]

    J. Yoon, J. Jordon, M. Schaar, Gain: Missing data imputation using generative adversarial nets, in: International conference on machine learning, PMLR, 2018, pp. 5689–5698

    2018

  42. [45]

    B. S. Van, H. C. Boshuizen, K. D. L, Multiple imputation of missing blood pressure covariates in survival analysis, Statistics in medicine 18 (6) (1999) 681–694

    1999

  43. [46]

    D. T. Neves, M. G. Naik, A. Proença, Sgain, wsgain-cp and wsgain- gp: Novel gan methods for missing data imputation, in: International Conference on Computational Science, 2021, pp. 98–113

    2021

  44. [47]

    T. N. Diogo, A. João, G. N. Marcel, J. P. Alberto, P. Fabian, From missingdataimputationtodatageneration,JournalofComputational Science 61 (2022) 101640

    2022

  45. [48]

    <Siyi Li et al.>:Preprint submitted to ElsevierPage 11 of 13 <TWVis>

    N.Park,M.Mohammadi,K.Gorde,S.Jajodia,H.Park,Y.Kim,Data synthesis based on generative adversarial networks, Proceedings of the VLDB Endowment (2018). <Siyi Li et al.>:Preprint submitted to ElsevierPage 11 of 13 <TWVis>

    2018

  46. [49]

    N. V. Chawla, K. W. Bowyer, L. O. Hall, W. P. Kegelmeyer, Smote: syntheticminorityover-samplingtechnique,Journalofartificialintel- ligence research 16 (2002) 321–357

    2002

  47. [50]

    Mohamed, E

    E. Mohamed, E. Omar, A.-M. Somaya, T. Hamid, Image generation: A review, Neural Processing Letters 54 (5) (2022) 4609–4646

    2022

  48. [51]

    Breiman, Random forests, Machine learning 45 (2001) 5–32

    L. Breiman, Random forests, Machine learning 45 (2001) 5–32. <Siyi Li et al.>:Preprint submitted to ElsevierPage 12 of 13 <TWVis> Supplementary Materials Figure A1: The military formation in Pit No. 1. <Siyi Li et al.>:Preprint submitted to ElsevierPage 13 of 13

    2001