Social Biases in Knowledge Representations of Wikidata separates Global North from Global South

Aditya Soni; Animesh Mukherjee; Paramita Das; Sai Keerthana Karnam

REVIEW 4 major objections 6 minor 59 references

Social Biases in Knowledge Representations of Wikidata separates Global North from Global South

T0 review · 4 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Bias patterns in Wikidata occupation predictions separate the Global North from the Global South.

desk verdict A valuable 21-geography bias audit undermined by an unnormalized aggregation step that likely makes the Global North/South split an artifact of Wikidata coverage volume rather than bias composition. read the letter →

arxiv 2505.02352 v1 pith:XSCTTNJY submitted 2025-05-05 cs.IR cs.AI

Paramita Das , Sai Keerthana Karnam , Aditya Soni , Animesh Mukherjee This is my paper · ORCID

classification cs.IRcs.AI

keywords knowledgegraphlinkpredictionWikidatasocialbiasfairnessgenderageGlobalNorth–GlobalSouthdivide

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that social biases in link prediction on Wikidata are not distributed randomly across the world: when occupations are labelled as male- or female-biased (and young- or old-biased) by standard fairness metrics, the per-country pattern of labels clusters into two groups that closely track the familiar economic and cultural divide between the Global North and the Global South. The authors build a framework, AuditLP, that hides a person's occupation from the graph, learns entity embeddings with four different knowledge-graph embedding algorithms, and uses a classifier to predict the hidden occupation links. The framework then classifies each occupation in each of 21 geographies into one of three bias categories for gender and for age. Aggregating these labels into a five-dimensional per-geography vector and running spectral clustering yields the North–South split, and the split appears for all four algorithms. If the claim holds, it means that the structural bias of a widely used open knowledge graph carries a fingerprint of global socio-economic divisions into downstream AI systems.

What carries the argument

The machinery has three linked parts. First, AuditLP: hide the occupation links of half the humans (stratified by gender), train a knowledge-graph embedding model, and feed head–relation–tail concatenations of the learned vectors into an MLP classifier to score whether each hidden link exists. Second, the labelling rule: using equal-opportunity and equalized-odds fairness criteria, each occupation is assigned to one of three categories according to differences between male and female (or young and old) true-positive and false-positive rates, with the threshold taken as the mean minus the standard deviation of the rate differences. Third, the geographic signature: each of the 21 countries is represented by a five-dimensional vector whose entries count how many of its occupations fall into the bias categories; these vectors are then clustered by hierarchical spectral clustering, and the resulting clusters are compared with global socio-economic indicators. The decisive move is that the same two-cluster separation emerges from four structurally different embedding models, which the paper takes as evidence that the pattern is a property of the data and task, not of a single learning algorithm.

What would settle it

Recompute each geography's five-dimensional vector after dividing every entry by the country's total number of occupations (or number of human entities), then re-run the spectral clustering. If the clean Global North/Global South split does not survive this normalization, the reported partition is an artifact of data volume and curation effort rather than of qualitative bias differences.

Watch

Extended reading notes

Core claim

The central discovery is that the bias profile of Wikidata's knowledge representations contains a global geographic signal: the vector of how many occupations fall into each of the five bias categories reproduces the Global North/Global South partition. The paper demonstrates this for two sensitive attributes, gender and age, across 21 geographies and four embedding models, and checks the clusters against country-level attributes such as GDP per capita, Human Development Index, Gender Gap Index, Gini coefficient, individualism, and cultural distance from the United States. It further shows that some occupations are oppositely biased in the two blocs: intellectual and white-collar occupations tend to be male-biased in the Global North and female-biased in the Global South, while physically demanding sports occupations show the reverse pattern. The paper interprets this as evidence that the bias in the knowledge graph and its embeddings is not merely algorithmic noise but mirrors real socio-economic differences in who gets recorded and how occupations are gendered.

Load-bearing premise

The load-bearing premise is that adding up per-occupation bias indicators within each country measures qualitative differences in bias, rather than merely reflecting how many occupations and people Wikidata happens to contain for richer, more heavily edited countries.

Editorial extensions

If this is right

Because the split appears for all four embedding model families, any of these families used for knowledge-graph completion will propagate a geographically structured gender or age bias into downstream applications such as question answering and language-model pretraining.
Occupations labelled male-biased in the Global North can be female-biased in the Global South and vice versa, so fairness interventions trained on one region may be wrong for the other.
Country-level indicators such as GDP per capita, Human Development Index, Gender Gap Index, Gini coefficient, individualism, and cultural distance correlate with the clusters, so the bias pattern can serve as a proxy signal of a country's position in the global economic order.
The framework labels occupations as age-biased as well as gender-biased for every geography, and the age-based clusters also follow the North–South division.
Aggregate metrics that average over all countries can obscure the fact that individual countries are biased in opposite directions, so future bias audits should report geography-disaggregated results.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A direct test of the authors' interpretation would be to replace the raw five-dimensional count vectors with normalized vectors (for example, dividing by the number of occupations in each geography); if the clean North–South split does not survive normalization, the partition may be driven by data volume and curation effort rather than by qualitative differences in bias.
If the pattern generalizes beyond gender and age to race, ethnicity, or religion, then debiasing methods for knowledge-graph completion may need region-specific objectives rather than a single global fairness constraint.
Adding more geographies not in the current set, such as China, Southeast Asia, or Central Asia, would stress-test the universality of the claimed partition: the claim would gain force if those countries fall into the predicted clusters and lose force if they do not.
The authors' aggregate-level results show only slight overall differences between male and female true-positive rates, which suggests that an analysis that stops at global averages would miss the strong, opposite, and geographically structured biases the paper finds.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The genuinely new thing here is the scope: 21 geographies, four structurally different embedding families, and a hidden-edge link prediction setup with fairness metrics applied per occupation. That is real work, and a region-tagged dataset of this size could be useful to the field. The paper is also honest enough to note that rich countries have occupations in almost all categories while smaller ones don't. I believe them. The problem is that their own observation sinks the headline result.

The load-bearing step is Section 6.2. Each occupation gets a 5-dimensional indicator vector for which bias conditions it satisfies, and then these are summed per geography. That unnormalized sum conflates two things: the qualitative mix of biased versus neutral occupations, and the sheer number of occupations Wikidata has for that country. Spectral clustering on raw counts is dominated by vector magnitude. So the resulting clusters — the US, France, Germany on one side, Arabia, Nigeria, India on the other — may simply be "lots of Wikidata coverage" versus "less coverage." The Table 3 economic indicators do not remove the confound because GDP, HDI, and Wikidata curation effort are correlated. I'm not saying there is no real bias geography in Wikidata; I'm saying this analysis does not establish it.

There are smaller issues. The thresholds t1 and t2 are computed as mean minus/plus sigma of the very same TPR/FPR differences they are used to label. That is fitting the threshold to the data without significance testing or confidence intervals. The 0.01 neutrality cutoff is arbitrary. And the availability statement is contradictory: the abstract and footnote say all code and data are available, while the conclusion says they will release upon acceptance. That needs cleanup.

Credit where it's due: the experimental pipeline is carefully described, the choice of four KGE families is sensible, and the idea of validating clusters against external country attributes is good, even if the validation is undercut by the confound. This is not a desk-reject paper; it deserves a serious referee. But the central claim is not established until the geography vectors are normalized — say, by proportion of biased occupations or with a per-occupation control — and until the clustering is shown to survive on a coverage-matched sample. I would send it back for major revision, not accept it as is. For my own work I would not cite the headline finding yet, though I might cite the dataset if it becomes available and is better documented.

Referee Report

4 major / 6 minor

Summary. The paper introduces AuditLP, a framework that measures gender and age bias in link prediction over Wikidata triples for 21 geographies using four knowledge-graph embedding models (TransE, DistMult, CompGCN, GeKC). Occupations are labeled male/female-biased, young/old-biased, or neutral from TPR/FPR differences, and each geography is represented by a five-dimensional count vector summing these labels across occupations. Spectral clustering of these vectors is claimed to reproduce a Global North / Global South partition, supported by comparisons with country-level economic and cultural indicators and by lists of occupations with opposite bias between clusters. The paper also presents a new geography-specific Wikidata dataset and states that code and data are available.

Significance. If the central claim holds, the paper would make a notable contribution: it scales bias auditing of knowledge-graph link prediction to 21 geographies, shows consistency across four quite different embedding models, and connects algorithmic bias patterns to macro-level socioeconomic divisions. The authors' effort to validate clusters with external country attributes is a positive feature, as is the stated intent to release code and data. However, the load-bearing geographical representation is an unnormalized count vector, so the headline Global North / Global South separation is not currently established as reflecting bias composition rather than Wikidata coverage volume. The result is potentially interesting, but it needs substantial additional analysis before it can be accepted.

major comments (4)

[6.2] The geography-level representation is defined as the sum of occupation-level five-dimensional indicator vectors over all occupations in that geography. Because the entries are counts rather than proportions, geographies with more occupations have larger vector magnitudes in every category, and spectral clustering on raw count vectors is dominated by this magnitude difference. A high-coverage versus low-coverage split can therefore look like a Global North / Global South partition even if per-occupation bias probabilities were identical across countries. The paper itself notes in Section 6.1 that larger-geography datasets contain occupations in almost all categories while smaller geographies such as Arabia and Israel do not. The authors should normalize the geography vectors (for example, by the number of occupations or by L2 norm) and/or include a coverage-matched control, and then show that the GN/GS clustering still emerges. The country attributes in Table 3 do not resolve this confound, since GDP, HDI, and Wikidata curation volume are correlated.
[5, Eqs. (6)-(10)] The thresholds t1 and t2 used to label occupations as biased or neutral are derived from the mean and standard deviation of the same TPR/FPR differences that are subsequently used to produce the labels, and the neutrality cutoff of 0.01 is fixed. This means the number of biased occupations is determined by the variance of the data by construction, and the labels are not accompanied by significance tests or confidence intervals. Please report the actual threshold values for each geography and model, and provide a sensitivity analysis over these thresholds; otherwise the occupation labels, and hence the geography vectors that drive the clustering, are not shown to be robust.
[6.2, Table 2] The claimed agreement with the Global North / Global South partition is not quantified, and several clusters do not align cleanly with that partition. For example, in the TransE gender result, C1 includes Argentina, Russia, and Israel while C2 includes Mexico, New Zealand, and South Korea; in the CompGCN gender result, C1 includes Brazil and C2 includes several countries usually classified as Global North. The statement that clusters are 'broadly consistent' appears to be a post-hoc reading. Please report a quantitative agreement measure (e.g., adjusted Rand index or cluster purity) against a reference GN/GS labeling, and compare with a null model using permuted labels or count-only features.
[6.2, Table 3] The quantitative evidence consists of cluster-level averages of six country attributes, but no within-cluster variance, standard errors, or statistical tests are provided. The text uses phrases such as 'significantly different' without supporting inference. At minimum, report the per-country values or provide confidence intervals and a test (e.g., a permutation test) for the difference between the GN and GS clusters on each attribute.

minor comments (6)

[1] There is a typo in the introduction: 'DistMuslt' should read 'DistMult'.
[1] The framework is described as a 'noble framework'; 'novel' is the intended word in this context.
[1 and 7] The data availability statement is inconsistent: Section 1 says all code and data are made available, while Section 7 says the full dataset and code will be released upon acceptance. Please clarify which is the case.
[3, Figure 1] Figure 1 is referenced as showing entity and occupation counts, but the counts are not discussed numerically in the text and the figure is not included in the manuscript text; please add a clear description and ensure the figure is readable.
[5] The edge-hiding step uses a single random 50% split while maintaining the male-female ratio, but no random seeds or multiple runs are reported. Some indication of variance across splits would help establish that the classification results are stable.
[6.1 and Table 1] In the age rows of Table 1, the text appears to duplicate TPR_yn and FPR_ol instead of listing TPR_yn, TPR_ol, FPR_yn, and FPR_ol; please correct the notation.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the GN/GS partition is externally validated; the coverage-volume confound is a correctness risk, not a circular derivation.

full rationale

The paper's central claim is that spectral clustering of geography-level occupation-bias vectors yields clusters consistent with the Global North/South divide. The derivation chain is: hide occupation edges from KG embeddings; train TransE/DistMult/CompGCN/GeKC; classify held-out occupation links; compute TPR/FPR per sensitive group and threshold them (Eqs. 6–10) to label occupations as biased or neutral; sum per-occupation 5-dimensional indicator vectors to form a geography vector in Sec. 6.2; spectral-cluster those vectors; and only then compare clusters to external country attributes such as GDP, HDI, Gini, gender gap, cultural distance, and individualism in Table 3. No step fits a parameter to the Global North/South labels, and the external attributes are not used to construct the clusters, so the partition is not equivalent to its input by construction. The self-citation [14] is used only as background on embedding-phase bias amplification and is not the load-bearing premise of the partition claim. Equations (6)–(10) do calibrate thresholds on the same TPR/FPR distributions they categorize, but this is an internal operationalization of 'bias,' not a prediction of the geographical separation. The unnormalized summation in Sec. 6.2, combined with the paper's own admission that 'geographies with a larger number of triples, such as the USA, Germany, and France, contain occupations in almost all categories in contrast to smaller geographies such as Arabia, Israel, etc.,' raises a genuine validity concern: cluster separation may be driven by Wikidata coverage volume rather than qualitative bias composition. That is a robustness/correctness threat, not circularity, because the conclusion does not reduce to its inputs by definition or by a fitted parameter. I therefore find no circular step. This non-finding is consistent with the rule that methodological confounds belong under correctness risk rather than circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claim rests on a small set of hand-set design choices: data-derived bias thresholds, an arbitrary neutrality cutoff, manually defined age bins, an elbow-chosen cluster count, and unnormalized per-geography count vectors. No new entities are postulated. Most axioms are standard machine-learning assumptions, but the paper's interpretation that clusters reflect social rather than data-coverage differences is an unvalidated domain assumption.

free parameters (6)

TPR threshold t1 for gender = mu - sigma of per-occupation TPR_m - TPR_f
Calculated with mu - sigma in Section 5, used in Equations 6 and 8 to decide male- and female-biased occupations; this controls how many occupations are labeled biased and therefore shapes the country vectors.
FPR threshold t2 for gender = mu - sigma of FPR_m - FPR_f
Used in Equations 7 and 9 for FPR-based bias labels; same data-derived selection as t1.
Age thresholds t1 and t2 = mu + sigma of TPR/FPR differences (young vs old)
Section 5 switches to mu + sigma for age, the opposite rule from gender, with no justification; this changes the age-bias labels and the resulting clusters.
Cluster count k = chosen by elbow method, exact values not reported
Section 6.2 uses hierarchical spectral clustering with k set by the elbow method; no stability or bootstrap analysis is shown.
Neutrality cutoff = 0.01
Equation 10 treats absolute TPR/FPR differences below 0.01 as neutral for all geographies; arbitrary global constant.
Age group boundaries = young 19-45, old 60-90, 15-year gap
Section 3 defines age groups by hand; different boundaries would change every age-bias category.

assumptions (5)

domain assumption Binary gender is sufficient for the audit.
Section 3 excludes non-binary entities because they are extremely few in Wikidata; this is a modeling decision that removes part of the population from the fairness analysis.
domain assumption Per-geography Wikidata coverage can be compared as if it were a social measurement.
Section 3 builds each geography graph from country-of-citizenship links; differences in editor effort and data completeness are treated as properties of the society, not as a confound.
domain assumption Hiding 50% of occupation triples keeps occupation information out of the learned embeddings.
Section 5 removes occupation edges before training but other relations (e.g., position held, field of work) may still let the model infer occupations; no leakage test is reported.
domain assumption Threshold-based TPR/FPR differences identify true bias rather than sampling noise.
Section 5 and Equations 6-10 classify occupations with no confidence intervals, significance tests, or prevalence-based corrections, even for occupations with only a handful of people.
standard math Spectral clustering with elbow-selected k produces a meaningful partition.
Section 6.2 uses standard clustering; the assumption is that the elbow rule picks a stable, meaningful k, which is not demonstrated with silhouettes or bootstraps.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Social Biases in Knowledge Representations of Wikidata separates Global North from Global South." pith.science (2026). https://pith.science/paper/XSCTTNJY

@misc{pith2026250502352,
  author       = {Pith},
  title        = {Pith review of: Social Biases in Knowledge Representations of Wikidata separates Global North from Global South},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XSCTTNJY}},
  note         = {Machine review of arXiv:2505.02352}
}

read the original abstract

Knowledge Graphs have become increasingly popular due to their wide usage in various downstream applications, including information retrieval, chatbot development, language model construction, and many others. Link prediction (LP) is a crucial downstream task for knowledge graphs, as it helps to address the problem of the incompleteness of the knowledge graphs. However, previous research has shown that knowledge graphs, often created in a (semi) automatic manner, are not free from social biases. These biases can have harmful effects on downstream applications, especially by leading to unfair behavior toward minority groups. To understand this issue in detail, we develop a framework -- AuditLP -- deploying fairness metrics to identify biased outcomes in LP, specifically how occupations are classified as either male or female-dominated based on gender as a sensitive attribute. We have experimented with the sensitive attribute of age and observed that occupations are categorized as young-biased, old-biased, and age-neutral. We conduct our experiments on a large number of knowledge triples that belong to 21 different geographies extracted from the open-sourced knowledge graph, Wikidata. Our study shows that the variance in the biased outcomes across geographies neatly mirrors the socio-economic and cultural division of the world, resulting in a transparent partition of the Global North from the Global South.

Figures

Figures reproduced from arXiv: 2505.02352 by the authors.

**Figure 2.** Schematic for our experiments showing different steps in case of sensitive attribute gender– edge hiding, embedding [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

59 extracted references · 42 canonical work pages

[1]

Oshin Agarwal, Heming Ge, Siamak Shakeri, and Rami Al-Rfou. 2021. Knowledge Graph Based Synthetic Corpus Generation for Knowledge-Enhanced Language Model Pre-training. In Proceedings of NAACL. 3554–3565

work page 2021
[2]

Mehdi Ali, Max Berrendorf, Charles Tapley Hoyt, Laurent Vermue, Mikhail Galkin, Sahand Sharifzadeh, Asja Fischer, Volker Tresp, and Jens Lehmann. 2021. Bringing light into the dark: A large-scale evaluation of knowledge graph embedding models under a unified framework. IEEE Transactions on Pattern Analysis and Machine Intelligence 44, 12 (2021), 8825–8845

work page 2021
[3]

Sudhir Anand and Amartya Sen. 1994. Human Development Index: Methodology and Measurement. (1994)

work page 1994
[4]

Categories

Mario Arduini, Lorenzo Noci, Federico Pirovano, Ce Zhang, Yash Raj Shrestha, and Bibek Paudel. 2020. Adversarial learning for debiasing knowledge graph Social Biases in Knowledge Representations of Wikidata Conference’17, July 2017, Washington, DC, USA Categories TransE DistMult CompGCN GeKC (𝑇𝑃𝑅𝑚 > 𝑇𝑃𝑅𝑓)𝐺𝑁 , (𝑇𝑃𝑅𝑚 <𝑇𝑃𝑅𝑓)𝐺𝑆 athlete, businessperson, artist...

arXiv 2020
[5]

Ivana Balažević, Carl Allen, and Timothy M Hospedales. 2019. Hypernetwork knowledge graph embeddings. In International Conference on Artificial Neural Networks. Springer, 553–565

work page 2019
[6]

Nathan Bartley, Andres Abeliuk, Emilio Ferrara, and Kristina Lerman. 2021. Auditing algorithmic bias on twitter. In Proceedings of the 13th ACM Web Science Conference 2021. 65–73

work page 2021
[7]

Pablo Beytía, Pushkal Agarwal, Miriam Redi, and Vivek K Singh. 2022. Visual gender biases in wikipedia: A systematic evaluation across the ten most spoken languages. In ICWSM, Vol. 16. 43–54

work page 2022
[8]

Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, and Ok- sana Yakhnenko. 2013. Translating embeddings for modeling multi-relational data. Advances in neural information processing systems 26 (2013)

2013

Show all 59 references

[9]

Avishek Bose and William Hamilton. 2019. Compositional fairness constraints for graph embeddings. In International Conference on Machine Learning . PMLR, 715–724

2019
[10]

Styliani Bourli and Evaggelia Pitoura. 2020. Bias in Knowledge Graph Embed- dings. In 2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM). IEEE, 6–10

2020
[11]

Benjamin Cabrera, Björn Ross, Marielle Dado, and Maritta Heisel. 2018. The gender gap in Wikipedia talk pages. In Proceedings of the International AAAI Conference on Web and Social Media , Vol. 12

2018
[12]

Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, and Xiangnan He. 2023. Bias and debias in recommender system: A survey and future directions. ACM Transactions on Information Systems 41, 3 (2023), 1–39

2023
[13]

Michele Coscia and Michael Szell. 2021. Multilayer graph association rules for link prediction. In Proceedings of the International AAAI Conference on Web and Social Media, Vol. 15. 129–139

2021
[14]

Paramita Das, Sai Keerthana Karnam, Anirban Panda, Bhanu Prakash Reddy Guda, Soumya Sarkar, and Animesh Mukherjee. 2023. Diversity matters: Robustness of bias measurements in Wikidata. In Proceedings of the 15th ACM Web Science Conference 2023. 208–218. Conference’17, July 201...

2023
[15]

Fernando G De Maio. 2007. Income inequality measures. Journal of Epidemiology & Community Health 61, 10 (2007), 849–852

2007
[16]

Gianluca Demartini. 2019. Implicit bias in crowdsourced knowledge graphs. In Companion Proceedings of The 2019 World Wide Web Conference . 624–630

2019
[17]

Tim Dettmers, Pasquale Minervini, Pontus Stenetorp, and Sebastian Riedel. 2018. Convolutional 2d knowledge graph embeddings. InThirty-second AAAI conference on artificial intelligence

2018
[18]

Yupei Du, Qi Zheng, Yuanbin Wu, Man Lan, Yan Yang, and Meirong Ma. 2022. Understanding gender bias in knowledge base embeddings. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 1381–1395

2022
[19]

Joseph Fisher, Arpit Mittal, Dave Palfrey, and Christos Christodoulopoulos. 2020. Debiasing knowledge graph embeddings. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) . 7332–7345

2020
[20]

Joseph Fisher, Dave Palfrey, Christos Christodoulopoulos, and Arpit Mit- tal. 2020. Measuring Social Bias in Knowledge Graph Embeddings. arXiv:1912.02761 [cs.CL]

2020 arXiv
[21]

Xiao Huang, Jingyuan Zhang, Dingcheng Li, and Ping Li. 2019. Knowledge graph embedding based question answering. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining . 105–113

2019
[22]

Daphna Keidar, Mian Zhong, Ce Zhang, Yash Raj Shrestha, and Bibek Paudel
[23]

Stephan Klasen. 2019. What explains uneven female labor force participation levels and trends in developing countries? The World Bank Research Observer 34, 2 (2019), 161–197

2019
[24]

Angelie Kraft and Ricardo Usbeck. 2022. The Lifecycle of “Facts”: A Survey of Social Bias in Knowledge Graphs. In Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natur...

2022
[25]

Ajay Kumar, Shashank Sheshar Singh, Kuldeep Singh, and Bhaskar Biswas. 2020. Link prediction techniques, applications, and performance: A survey. Physica A: Statistical Mechanics and its Applications 553 (2020), 124289

2020
[26]

Preethi Lahoti, Krishna P Gummadi, and Gerhard Weikum. 2019. ifair: Learning individually fair data representations for algorithmic decision making. In 2019 ieee 35th international conference on data engineering (icde) . IEEE, 1334–1345

2019
[27]

Lorenzo Loconte, Nicola Di Mauro, Robert Peharz, and Antonio Vergari. 2024. How to turn your knowledge graph embeddings into generative models.Advances in Neural Information Processing Systems 36 (2024)

2024
[28]

Liu, Matthew E

Robert Logan, Nelson F. Liu, Matthew E. Peters, Matt Gardner, and Sameer Singh. 2019. Barack’s Wife Hillary: Using Knowledge Graphs for Fact-Aware Language Modeling. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 5962–5971

2019
[29]

Michael Muthukrishna, Adrian V Bell, Joseph Henrich, Cameron M Curtin, Alexander Gedranovich, Jason McInerney, and Braden Thue. 2020. Beyond Western, Educated, Industrial, Rich, and Democratic (WEIRD) psychology: Mea- suring and mapping scales of cultural and psychological dis...

2020
[30]

Ivana Načinović Braje, Maja Klindžić, and Lovorka Galetić. 2019. The role of individual variable pay in a collectivistic culture society: An evaluation.Economic research-Ekonomska istraživanja 32, 1 (2019), 1352–1372

2019
[31]

Dai Quoc Nguyen, Tu Dinh Nguyen, Dat Quoc Nguyen, and Dinh Phung. 2017. A novel embedding model for knowledge base completion based on convolutional neural network. arXiv preprint arXiv:1712.02121 (2017)

2017 arXiv
[32]

Maximilian Nickel, Lorenzo Rosasco, and Tomaso Poggio. 2016. Holographic embeddings of knowledge graphs. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 30

2016
[33]

Maximilian Nickel, Volker Tresp, and Hans-Peter Kriegel. 2011. A three-way model for collective learning on multi-relational data. In Icml

2011
[34]

Lemuel Ekedegwa Odeh. 2010. A comparative analysis of global north and global south economies. (2010)

2010
[35]

Esteban Ortiz-Ospina, Sandra Tzvetkova, and Max Roser. 2024. Women’s em- ployment. Our world in data (2024)

2024
[36]

Millenio Ramadizsa, Fariz Darari, and Werner Nutt. 2023. [Novel] Knowledge gap discovery: A case study of Wikidata. In The 4th Wikidata Workshop. https: //openreview.net/forum?id=nqDknjB9X8

2023
[37]

Clemens Reimann, Peter Filzmoser, and Robert G. Garrett. 2005. Background and threshold: critical comparison of methods of determination. Science of The Total Environment 346, 1 (2005), 1–16. https://doi.org/10.1016/j.scitotenv.2004.11.023

2005 doi
[38]

Andrea Rossi, Denilson Barbosa, Donatella Firmani, Antonio Matinata, and Paolo Merialdo. 2021. Knowledge graph embedding for link prediction: A comparative analysis. ACM Transactions on Knowledge Discovery from Data (TKDD) 15, 2 (2021), 1–49

2021
[39]

Apoorv Saxena, Aditay Tripathi, and Partha Talukdar. 2020. Improving multi-hop question answering over knowledge graphs using knowledge base embeddings. In Proceedings of the 58th annual meeting of the association for computational linguistics. 4498–4507

2020
[40]

Lena Schwertmann, Manoj Prabhakar Kannan Ravi, and Gerard De Melo. 2023. Model-agnostic bias measurement in link prediction. InFindings of the Association for Computational Linguistics: EACL 2023 . 1632–1648

2023
[41]

Zaina Shaik, Filip Ilievski, and Fred Morstatter. 2021. Analyzing race and country of citizenship bias in wikidata. arXiv preprint arXiv:2108.05412 (2021)

2021 arXiv
[42]

Chao Shang, Yun Tang, Jing Huang, Jinbo Bi, Xiaodong He, and Bowen Zhou
[43]

Harry Shomer, Wei Jin, Wentao Wang, and Jiliang Tang. 2023. Toward Degree Bias in Embedding-Based Knowledge Graph Completion. In Proceedings of the ACM Web Conference 2023. 705–715

2023
[44]

Théo Trouillon, Johannes Welbl, Sebastian Riedel, Éric Gaussier, and Guillaume Bouchard. 2016. Complex embeddings for simple link prediction. In International conference on machine learning . PMLR, 2071–2080

2016
[45]

Shikhar Vashishth, Soumya Sanyal, Vikram Nitin, and Partha Talukdar. 2019. Composition-based multi-relational graph convolutional networks.arXiv preprint arXiv:1911.03082 (2019)

2019 arXiv
[46]

Denny Vrandečić, Lydia Pintscher, and Markus Krötzsch. 2023. Wikidata: The making of. In Companion Proceedings of the ACM Web Conference 2023 . 615–624

2023
[47]

Claudia Wagner, David Garcia, Mohsen Jadidi, and Markus Strohmaier. 2015. It’s a man’s Wikipedia? Assessing gender inequality in an online encyclopedia. In Ninth international AAAI conference on web and social media

2015
[48]

Meihong Wang, Linling Qiu, and Xiaoli Wang. 2021. A Survey on Knowledge Graph Embeddings for Link Prediction. Symmetry 13, 3 (2021), 485

2021
[49]

Quan Wang, Zhendong Mao, Bin Wang, and Li Guo. 2017. Knowledge graph embedding: A survey of approaches and applications. IEEE transactions on knowledge and data engineering 29, 12 (2017), 2724–2743

2017
[50]

Yu Wang and Tyler Derr. 2022. Degree-Related Bias in Link Prediction. In 2022 IEEE International Conference on Data Mining Workshops (ICDMW) . IEEE, 757– 758

2022
[51]

Bingke Xu, Yue Cui, Zipeng Sun, Liwei Deng, and Kai Zheng. 2021. Fair represen- tation learning in knowledge-aware recommendation. In 2021 IEEE international conference on big knowledge (ICBK) . IEEE, 385–392

2021
[52]

Muhammad Yahya, John G Breslin, and Muhammad Intizar Ali. 2021. Semantic web and knowledge graphs for industry 4.0. Applied Sciences 11, 11 (2021), 5110

2021
[53]

Bishan Yang, Scott Wen-tau Yih, Xiaodong He, Jianfeng Gao, and Li Deng. 2015. Embedding Entities and Relations for Learning and Inference in Knowledge Bases. In Proceedings of the International Conference on Learning Representations (ICLR) 2015

2015
[54]

Charles Chuankai Zhang and Loren Terveen. 2021. Quantifying the Gap: A Case Study of Wikidata Gender Disparities. In 17th International Symposium on Open Collaboration. 1–12

2021
[55]

Wen Zhang, Bibek Paudel, Wei Zhang, Abraham Bernstein, and Huajun Chen
[56]

Zhanqiu Zhang, Jianyu Cai, Yongdong Zhang, and Jie Wang. 2020. Learning hierarchy-aware knowledge graph embeddings for link prediction. InProceedings of the AAAI conference on artificial intelligence , Vol. 34. 3065–3072

2020
[58]

In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining

Interaction embeddings for prediction and explanation in knowledge graphs. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining. 96–104
[2019]

In Proceedings of the AAAI conference on artificial intelligence , Vol

End-to-end structure-aware convolutional networks for knowledge base completion. In Proceedings of the AAAI conference on artificial intelligence , Vol. 33. 3060–3067
[2021]

arXiv preprint arXiv:2109.10697 (2021)

Towards Automatic Bias Detection in Knowledge Graphs. arXiv preprint arXiv:2109.10697 (2021)

2021 arXiv

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [1]

Oshin Agarwal, Heming Ge, Siamak Shakeri, and Rami Al-Rfou. 2021. Knowledge Graph Based Synthetic Corpus Generation for Knowledge-Enhanced Language Model Pre-training. In Proceedings of NAACL. 3554–3565

work page 2021

[2] [2]

Mehdi Ali, Max Berrendorf, Charles Tapley Hoyt, Laurent Vermue, Mikhail Galkin, Sahand Sharifzadeh, Asja Fischer, Volker Tresp, and Jens Lehmann. 2021. Bringing light into the dark: A large-scale evaluation of knowledge graph embedding models under a unified framework. IEEE Transactions on Pattern Analysis and Machine Intelligence 44, 12 (2021), 8825–8845

work page 2021

[3] [3]

Sudhir Anand and Amartya Sen. 1994. Human Development Index: Methodology and Measurement. (1994)

work page 1994

[4] [4]

Categories

Mario Arduini, Lorenzo Noci, Federico Pirovano, Ce Zhang, Yash Raj Shrestha, and Bibek Paudel. 2020. Adversarial learning for debiasing knowledge graph Social Biases in Knowledge Representations of Wikidata Conference’17, July 2017, Washington, DC, USA Categories TransE DistMult CompGCN GeKC (𝑇𝑃𝑅𝑚 > 𝑇𝑃𝑅𝑓)𝐺𝑁 , (𝑇𝑃𝑅𝑚 <𝑇𝑃𝑅𝑓)𝐺𝑆 athlete, businessperson, artist...

arXiv 2020

[5] [5]

Ivana Balažević, Carl Allen, and Timothy M Hospedales. 2019. Hypernetwork knowledge graph embeddings. In International Conference on Artificial Neural Networks. Springer, 553–565

work page 2019

[6] [6]

Nathan Bartley, Andres Abeliuk, Emilio Ferrara, and Kristina Lerman. 2021. Auditing algorithmic bias on twitter. In Proceedings of the 13th ACM Web Science Conference 2021. 65–73

work page 2021

[7] [7]

Pablo Beytía, Pushkal Agarwal, Miriam Redi, and Vivek K Singh. 2022. Visual gender biases in wikipedia: A systematic evaluation across the ten most spoken languages. In ICWSM, Vol. 16. 43–54

work page 2022

[8] [8]

Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, and Ok- sana Yakhnenko. 2013. Translating embeddings for modeling multi-relational data. Advances in neural information processing systems 26 (2013)

2013