Pith. sign in

REVIEW 3 major objections 4 minor 21 references

Graph Node Embeddings using Domain-Aware Biased Random Walks

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper claims that domain-aware biased random walks—weighting graph edges by what they mean—produce node embeddings that match or exceed structure-only approaches in downstream classification.

desk verdict The AIFB 99.86% result is likely label leakage, not a general semantic-embedding win, but the framework is honest and deserves careful referee time. read the letter →

arxiv 1908.02947 v1 pith:O4HBVJF6 submitted 2019-08-08 cs.LG cs.SIstat.ML

Sourav Mukherjee , Tim Oates , Ryan Wright This is my paper · ORCID
classification cs.LGcs.SIstat.ML
keywords graphembeddingsbiasedrandomwalksRDFgraphsnodedomain-awarewalksamplingword2vecsemantic
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

Graph-embedding methods typically generate vectors from random walks that are blind to what the graph's nodes and edges mean. This paper argues that exploiting domain semantics—the labels on nodes and edges and knowledge of the downstream prediction task—can produce better embeddings with no extra global computation. Its framework expresses each bias as an edge-weighting function: during a walk, an outgoing edge is chosen with probability proportional to its weight, so semantically relevant substructures appear in walk corpora more often. On the AIFB benchmark, hand-designed domain-aware weights lift k-Nearest-Neighbor classification from 77.42% for the best domain-independent bias to 99.86%; on the BGS benchmark, the best domain-aware strategy matches the best structural strategy while using only local information. The intended upshot is that semantic priors, not just topology, should be a first-class input to node embedding.

What carries the argument

The load-bearing object is the edge-weighting function $w: E \to \mathbb{R}_{>0}$: at a node with outgoing edges $e_1,\dots,e_k$, the next edge is selected with probability $w(e_i)/\sum_j w(e_j)$. Domain-aware strategies instantiate this function using locally visible information—edge labels such as 'affiliation' and 'member', or the type of the node an edge leads to—so that task-relevant structures are over-sampled in the generated walk corpus and irrelevant ones are under-sampled. It carries the argument because it converts qualitative domain knowledge directly into the walk distribution that feeds word2vec, the neural language-modeling algorithm that turns the walk corpus into dense vectors, and because it needs no graph-wide statistics such as label frequencies or node in-degrees.

What would settle it

Run k-NN classification on AIFB and BGS after deleting, or holding out, the 'affiliation'/'member' edges and 'hasLithogenesis' edges incident to test nodes before generating walks; if the domain-aware strategies fall to parity with uniform random walks, the gain came from walking along the answer-key edges.

Watch

Extended reading notes

Core claim

The central claim, on the paper's own terms, is that domain-aware biased random walks yield node embeddings that are simple to implement and achieve equal or greater accuracy than domain-independent approaches on downstream machine learning tasks. The mechanism is a family of edge-weighting functions that assign large weights to edges whose labels or endpoint types correspond to the target concept (for example, 'affiliation' and 'member' edges between people and research groups) and small weights to edges that lead to misleading nodes (such as publication nodes with many external co-authors). The reported numbers put the claim in concrete form: on the AIFB and BGS benchmark graphs, the combined AIFB weight function reaches 99.86% accuracy, far above the 77.42% of the best structure-only bias; on BGS, the best domain-aware function lands at 90.07% under Skip-gram, between the uniform baseline and the best structural frequency-based bias, while avoiding graph-wide frequency computations. The authors do not claim a universal win: BGS's best domain-independent strategy remains slightly ahead, and they attribute the gap to their own limited geological expertise.

Load-bearing premise

The evaluation assumes the graphs still contain the edges that directly encode the classification target for both training and test nodes; if those label-encoding edges were removed or held out for test instances, the reported accuracy advantage of the domain-aware walks would likely disappear.

Editorial extensions

If this is right

  • Practitioners can improve classification accuracy on RDF knowledge graphs by hand-designing edge weights from a few semantic cues, without computing global graph statistics.
  • Because the biases use only local information, walk generation can be parallelized or cached across overlapping neighborhoods, unlike frequency-based alternatives.
  • In domains where the label concept is encoded as graph structure (affiliation edges, lithogenesis edges), emphasizing those edges can raise accuracy dramatically; the effect will be strongest when such edges are present for test instances.
  • Negative biases can be as important as positive ones: deliberately avoiding publication nodes on AIFB improved accuracy to 91.56%, and adding the positive affiliation bias pushed it to 99.86%.
  • On less familiar domains, domain-aware strategies may only match, not beat, structural baselines without sustained domain expertise.

Reading between the lines

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

  • An implied test the paper does not run: remove or hold out the label-encoding edges for test nodes before generating walks. If classification accuracy collapses toward the uniform baseline, the measured advantage is attributable to the walker seeing the answer key, not to semantic awareness per se.
  • The framework can be read as a recipe for automated search: because each strategy is a small weighting function, one could optimize hyperparameters or compose primitive semantic biases through local search, something the paper leaves for future work.
  • The conflicting intuitions around co-authorship suggest semantic biases need not be monotone; a structure that helps in one graph (co-authorship ties) can hurt in another when communities are mixed, so the practical guidance is to test both positive and negative biases.
  • The BGS result hints that for non-expert users, global inverse-frequency biases remain a strong default; domain-aware weighting pays off most when users can name the exact relation that encodes the target.
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.

Referee Report

3 major / 4 minor

Summary. The paper proposes a framework for domain-aware biased random walks for RDF graph embeddings. The central idea is that semantic knowledge about node and edge types can be encoded as edge-weighting functions, and that the resulting weighted random walks, fed through word2vec, produce embeddings that improve downstream classification. The method is evaluated on two RDF datasets, AIFB (predicting the research group of a person) and BGS (predicting lithogenesis of a rock unit), using k-NN classification and compared against uniform random walks and standard domain-independent biasing strategies such as predicate frequency, inverse predicate frequency, and object frequency. The paper reports near-perfect accuracy (99.86%) for its best AIFB strategy, while its best BGS strategy is somewhat below the best domain-independent baseline.

Significance. If the results were valid, the framework would be a valuable practical contribution: the proposed weighting functions are simple, require only local neighborhood information rather than graph-wide statistics, and are clearly specified. The paper is also transparent about the BGS results in Section 4.3. However, the empirical evidence for the central claim is not sound. The headline AIFB result is confounded by direct label encoding, and the BGS results contradict the abstract's claim of 'equal or greater accuracy' compared with domain-independent approaches. Because the core claim rests on these experiments, the significance of the paper as it stands is limited.

major comments (3)
  1. [Section 3.2.1, Table 1] The headline AIFB result is invalidated by label leakage. Section 3.2.1 ('Strategy 1') states that the AIFB graph 'explicitly includes nodes of type research group, as well as edges labeled affiliation from persons to research groups and edges labeled member from research groups to persons for the training and test instances.' The classification label for a person is exactly the research group to which that person belongs. Algorithms 1 and 4 assign high weights to precisely these target-defining edges (whigh = 10 and whigh = 100 in Table 2). A random walk that preferentially traverses affiliation and member edges will place a test person's embedding close to the corresponding group node and to other members of the same group, so the k-NN classifier (k = 4) trivially recovers labels that are already present as graph structure. The 99.86% accuracy in Table 1 therefore does not establish that domain-aware semantics improve general embeddings; it shows that weighting the label-encoding subgraph maximally lets the classifier read the label off the graph. A comparison against domain-independent baselines under this setup is not evidence for the abstract's claim. The authors should remove or hold out the affiliation/member edges for test instances, or use a target-relation-free graph, before drawing conclusions.
  2. [Section 4.3, Table 3] The abstract and conclusion claim 'equal or greater accuracy' compared with domain-independent approaches, but the BGS results do not support this claim. Table 3 shows BGS-Weight-Function-1 at 84.21% (CBOW) and 90.07% (Skip-gram), while Inverse Predicate Frequency reaches 88.45% and 93.10% respectively. The best domain-specific strategy is lower than the best domain-independent strategy in both settings. Section 4.3 acknowledges this only as being 'in the ballpark' and states that the authors have not yet defined a BGS strategy that outperforms the best domain-independent strategy. The paper should either qualify the central claim explicitly or restrict it to 'equal in some cases and lower in others, with reduced computational cost.' As written, the conclusion overstates the evidence.
  3. [Section 4.2, Tables 2 and 4] The experimental comparison is not protected against overfitting. The paper states in Section 4.2 that 'we have tuned hyperparameters manually; use of automated hyperparameter tuning is a possible extension of this work,' and Tables 2 and 4 list many tuned values. No validation set or cross-validation procedure is described. With free parameters for each weighting function (wlow, w, whigh in various combinations), the k value in k-NN, and the word2vec settings (window, epochs, negative samples, dimensionality), the reported single-point accuracies may reflect tuning on the test set rather than a systematic advantage. The authors should provide a validation protocol and report variance across runs or folds.
minor comments (4)
  1. [Algorithm 9] The pseudocode for BGS-Weight-Function-5 does not match its description. In Algorithm 9, the final assignment 'weight← w' appears inside the else-branch and executes after the inner if, so for hasLithogenesis edges any value set to wlow or whigh is immediately overwritten by w. As printed, the algorithm assigns whigh to broader edges and w to everything else, never returning wlow or whigh for hasLithogenesis edges.
  2. [Section 4.1] No code or data release is mentioned. Since the framework is simple and the paper emphasizes ease of implementation, providing a link to the code would help reproducibility of Tables 1 and 3.
  3. [Table 1 and surrounding text] The text says that Skip-gram word2vec reduces measured accuracy for AIFB, but no skip-gram numbers are reported for AIFB. Either report the numbers or state that they are omitted for brevity.
  4. [Section 4.1] The choice of k=4 for AIFB and k=10 for BGS is justified by reproducing baseline results from Ristoski and Paulheim, but the paper does not show that reproduction. Including the reproduced uniform-walk baselines would make the comparison clearer.

Circularity Check

2 steps flagged · score 7.0 of 10

AIFB's 99.86% result is forced by construction: the domain-aware strategy upweights the exact affiliation/member edges that encode the classification label for training and test instances.

  1. self definitional [Section 3.2.1, Strategy 1 and Strategy 4; Table 1]
    "was used by Bloehdorn and Sure (2007) to predict the research group that any person is affiliated with. ... the RDF graph for AIFB ... explicitly includes nodes ... of type 'research group', as well as edges ... labeled 'affiliation' from persons to research groups and edges labeled 'member' from research groups to persons for the training and test instances. ... any edge whose label is 'affiliation' or 'member' is assigned weight whigh."

    The target of the AIFB classification is research-group affiliation, and the graph already contains affiliation/member edges connecting every person, including test instances, to their true group. Algorithm 4 sets the highest weight (whigh=100, Table 2) on exactly those edges. Random walks from a test person are therefore dominated by traversals along the test node's own label edge, placing its embedding near other members of the same research group; kNN then recovers the label from the graph structure. The 'prediction' is a re-reading of the label edges that the strategy was deliberately built to favor, so the 99.86% figure is not evidence that domain-aware semantics improve general embeddings.

  2. self definitional [Section 3.2.2, Strategy 1; Table 3]
    "was used in machine learning by de Vries (2013) to predict lithogenesis (method of formation) types of named rock units. ... the RDF graph for BGS ... contains nodes ... representing lithogenesis types, and edges ... labeled 'hasLithogenesis' associating rock types to lithogenesis types. ... our strategy ... is shown in Algorithm 5, where any edge labeled 'hasLithogenesis' is assigned weight whigh."

    Same reduction: the BGS classification label (lithogenesis type) is encoded in hasLithogenesis edges, and BGS-Weight-Function-1 upweights exactly those edges during walk generation. The resulting embeddings place rock units near their own label nodes, so k-NN can recover labels from the graph structure rather than from learned semantic regularities. The comparison to uniform or other structural baselines is confounded by this direct label accessibility. Additionally, Table 3 shows this strategy (90.07% Skip-gram) does not match Inverse Predicate Frequency (93.10%), so it does not support the abstract's 'equal or greater accuracy' claim.

full rationale

The circularity is confined to evaluation but load-bearing: the abstract's strongest evidence is the AIFB 99.86% vs 77.42% comparison, and the paper's own text shows that the winning weighting function favors the exact edges that encode the target for training and test nodes. That is a reduction by construction of the headline result, not an independent demonstration that semantic information improves embeddings. No self-citation or imported uniqueness theorem is involved. BGS exhibits the same label-edge leakage and, moreover, the best domain-specific strategy does not beat Inverse Predicate Frequency, so the 'equal or greater accuracy' claim is not established. Score reflects a central empirical claim that is forced by the design of the weighting function, while leaving the general framework itself non-circular.

Assumptions & free parameters 8 free parameters · 4 assumptions · 0 invented entities

The central method relies on a set of hand-chosen weights (hyperparameters) and on the domain assumption that the graph contains the target relation as ordinary edges. No new physical or mathematical entities are introduced. The hyperparameters are tuned on the same data used for evaluation, and the target-relation presence is the key premise that makes the AIFB result possible.

free parameters (8)
  • wlow, whigh for AIFB-Weight-Function-1, 2, 3 = 0.1, 10
    Manually tuned hyperparameters; no sensitivity analysis or validation split reported.
  • wlow, w, whigh for AIFB-Weight-Function-4 = 0.1, 10, 100
    Manually tuned; the paper states manual tuning is necessary.
  • wlow, whigh for BGS-Weight-Function-1 = 0.1, 10
    Manually tuned to upweight hasLithogenesis edges.
  • wlow, whigh for BGS-Weight-Function-2 = 0.001, 1
    Manually tuned to effectively remove hasLithogenesis edges.
  • wlow, w, whigh for BGS-Weight-Function-3 = 0.1, 10, 100
    Manually tuned; strategy weights concept hierarchy edges.
  • wlow, whigh for BGS-Weight-Function-4 = 0.1, 1
    Manually tuned to avoid inScheme edges.
  • k in k-NN classifier = 4 for AIFB, 10 for BGS
    The paper states these values were chosen so the implementation would reproduce the uniform-walk baseline from Ristoski and Paulheim (2016), i.e., tuned against a reported result rather than an independent validation set.
  • word2vec settings (window, epochs, negative samples, dimensionality) = window 5, epochs 10, negative 25, dim 200
    Values copied from prior work; not treated as free in the reported experiments.
assumptions (4)
  • domain assumption Random walks converted to word2vec embeddings preserve semantic proximity of nodes in the graph.
    This is the DeepWalk/RDF2Vec premise on which the entire method is built; used throughout Sections 3.1 and 3.2 without further justification.
  • domain assumption k-NN classification accuracy on a single train/test split is a valid measure of embedding quality.
    Section 4 uses only k-NN accuracy, with no repeated splits, no other tasks, and no statistical tests.
  • ad hoc to paper The graph used for embedding may include the edges and nodes that constitute the classification target (affiliation/member in AIFB, hasLithogenesis in BGS).
    The paper explicitly states these target-encoding edges are present for training and test instances (Sections 3.2.1 and 3.2.2). The headline accuracies depend on this premise; if target edges were held out, the method's advantage would likely vanish.
  • ad hoc to paper Manual hyperparameter selection on the test set is an acceptable experimental practice.
    The paper says hyperparameters were tuned manually and does not use a separate validation set; this is stated in Section 4.2 and the hyperparameter tables.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Graph Node Embeddings using Domain-Aware Biased Random Walks." pith.science (2026). https://pith.science/paper/O4HBVJF6

@misc{pith2026190802947,
  author       = {Pith},
  title        = {Pith review of: Graph Node Embeddings using Domain-Aware Biased Random Walks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O4HBVJF6}},
  note         = {Machine review of arXiv:1908.02947}
}
read the original abstract

The recent proliferation of publicly available graph-structured data has sparked an interest in machine learning algorithms for graph data. Since most traditional machine learning algorithms assume data to be tabular, embedding algorithms for mapping graph data to real-valued vector spaces has become an active area of research. Existing graph embedding approaches are based purely on structural information and ignore any semantic information from the underlying domain. In this paper, we demonstrate that semantic information can play a useful role in computing graph embeddings. Specifically, we present a framework for devising embedding strategies aware of domain-specific interpretations of graph nodes and edges, and use knowledge of downstream machine learning tasks to identify relevant graph substructures. Using two real-life domains, we show that our framework yields embeddings that are simple to implement and yet achieve equal or greater accuracy in machine learning tasks compared to domain independent approaches.

Figures

Figures reproduced from arXiv: 1908.02947 by the authors.

Figure 1
Figure 1. Node embedding framework for RDF graphs. [PITH_FULL_IMAGE:figures/full_fig_p016_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 12 canonical work pages

  1. [1]

    Linked data - the story so far

    Christian Bizer, Tom Heath, and Tim Berners-Lee. Linked data - the story so far. Int. J. Semantic Web Inf. Syst., 5: 0 1--22, 2009

    work page 2009

  2. [2]

    Kernel methods for mining instance data in ontologies

    Stephan Bloehdorn and York Sure. Kernel methods for mining instance data in ontologies. In Karl Aberer, Key - Sun Choi, Natasha Fridman Noy, Dean Allemang, Kyung - Il Lee, Lyndon J. B. Nixon, Jennifer Golbeck, Peter Mika, Diana Maynard, Riichiro Mizoguchi, Guus Schreiber, and Philippe Cudr \' e - Mauroux, editors, The Semantic Web, 6th International Seman...

    work page 2007

  3. [3]

    Webb, Dragos D

    Longbing Cao, Chengqi Zhang, Thorsten Joachims, Geoffrey I. Webb, Dragos D. Margineantu, and Graham Williams, editors. Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Sydney, NSW, Australia, August 10-13, 2015 , 2015. ACM . ISBN 978-1-4503-3664-2. URL http://dl.acm.org/citation.cfm?id=2783258

    work page 2015

  4. [4]

    Deep neural networks for learning graph representations

    Shaosheng Cao, Wei Lu, and Qiongkai Xu. Deep neural networks for learning graph representations. In Dale Schuurmans and Michael P. Wellman, editors, Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, February 12-17, 2016, Phoenix, Arizona, USA. , pages 1145--1152. AAAI Press, 2016. ISBN 978-1-57735-760-5. URL http://www.aaai.org/ocs/...

    work page 2016

  5. [5]

    Aggarwal, and Thomas S

    Shiyu Chang, Wei Han, Jiliang Tang, Guo - Jun Qi, Charu C. Aggarwal, and Thomas S. Huang. Heterogeneous network embedding via deep architectures. In DBLP:conf/kdd/2015 , pages 119--128. ISBN 978-1-4503-3664-2. doi:10.1145/2783258.2783296. URL http://doi.acm.org/10.1145/2783258.2783296

    arXiv 2015

  6. [6]

    Biased graph walks for RDF graph embeddings

    Michael Cochez, Petar Ristoski, Simone Paolo Ponzetto, and Heiko Paulheim. Biased graph walks for RDF graph embeddings. In Rajendra Akerkar, Alfredo Cuzzocrea, Jannong Cao, and Mohand - Said Hacid, editors, Proceedings of the 7th International Conference on Web Intelligence, Mining and Semantics, WIMS 2017, Amantea, Italy, June 19-22, 2017 , pages 21:1--2...

    arXiv 2017

  7. [7]

    A fast approximation of the weisfeiler-lehman graph kernel for RDF data

    Gerben Klaas Dirk de Vries. A fast approximation of the weisfeiler-lehman graph kernel for RDF data. In Hendrik Blockeel, Kristian Kersting, Siegfried Nijssen, and Filip Zelezn \' y , editors, Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2013, Prague, Czech Republic, September 23-27, 2013, Proceedings, Part I , vo...

    work page doi:10.1007/978-3-642-40988-2_39 2013

  8. [8]

    Substructure counting graph kernels for machine learning from RDF data

    Gerben Klaas Dirk de Vries and Steven de Rooij. Substructure counting graph kernels for machine learning from RDF data. J. Web Sem., 35: 0 71--84, 2015. doi:10.1016/j.websem.2015.08.002. URL https://doi.org/10.1016/j.websem.2015.08.002

    work page doi:10.1016/j.websem.2015.08.002 2015

Show all 21 references
  1. [9]

    node2vec: Scalable feature learning for networks

    Aditya Grover and Jure Leskovec. node2vec: Scalable feature learning for networks. In Balaji Krishnapuram, Mohak Shah, Alexander J. Smola, Charu C. Aggarwal, Dou Shen, and Rajeev Rastogi, editors, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discove...

    2016

  2. [10]

    Deep convolutional networks on graph-structured data

    Mikael Henaff, Joan Bruna, and Yann LeCun. Deep convolutional networks on graph-structured data. CoRR, abs/1506.05163, 2015. URL http://arxiv.org/abs/1506.05163

    2015 arXiv

  3. [11]

    SNAP Datasets : Stanford large network dataset collection

    Jure Leskovec and Andrej Krevl. SNAP Datasets : Stanford large network dataset collection. http://snap.stanford.edu/data, June 2014

    2014

  4. [12]

    McCormick

    C. McCormick. Word2vec tutorial - the skip-gram model. Retrieved from http://www.mccormickml.com, April 2016

    2016

  5. [13]

    Efficient estimation of word representations in vector space

    Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. Efficient estimation of word representations in vector space. CoRR, abs/1301.3781, 2013. URL http://arxiv.org/abs/1301.3781

    2013 arXiv

  6. [14]

    Learning convolutional neural networks for graphs

    Mathias Niepert, Mohamed Ahmed, and Konstantin Kutzkov. Learning convolutional neural networks for graphs. In Maria - Florina Balcan and Kilian Q. Weinberger, editors, Proceedings of the 33nd International Conference on Machine Learning, ICML 2016, New York City, NY, USA, June...

    2016

  7. [15]

    Deepwalk: online learning of social representations

    Bryan Perozzi, Rami Al - Rfou, and Steven Skiena. Deepwalk: online learning of social representations. In Sofus A. Macskassy, Claudia Perlich, Jure Leskovec, Wei Wang, and Rayid Ghani, editors, The 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining...

    2014

  8. [16]

    Software Framework for Topic Modelling with Large Corpora

    Radim R eh u r ek and Petr Sojka. Software Framework for Topic Modelling with Large Corpora . In Proceedings of the LREC 2010 Workshop on New Challenges for NLP Frameworks , pages 45--50, Valletta, Malta, May 2010. ELRA. http://is.muni.cz/publication/884893/en

    2010

  9. [17]

    o tzsch, Freddy L \' e cu \' e , Fabian Fl \

    Petar Ristoski and Heiko Paulheim. Rdf2vec: RDF graph embeddings for data mining. In Paul T. Groth, Elena Simperl, Alasdair J. G. Gray, Marta Sabou, Markus Kr \" o tzsch, Freddy L \' e cu \' e , Fabian Fl \" o ck, and Yolanda Gil, editors, The Semantic Web - ISWC 2016 - 15th I...

    2016 doi

  10. [18]

    o tzsch, Freddy L \' e cu \' e , Fabian Fl \

    Petar Ristoski, Gerben Klaas Dirk de Vries, and Heiko Paulheim. A collection of benchmark datasets for systematic evaluations of machine learning on the semantic web. In Paul T. Groth, Elena Simperl, Alasdair J. G. Gray, Marta Sabou, Markus Kr \" o tzsch, Freddy L \' e cu \' e...

    2016

  11. [19]

    Adoption of the linked data best practices in different topical domains

    Max Schmachtenberg, Christian Bizer, and Heiko Paulheim. Adoption of the linked data best practices in different topical domains. In Peter Mika, Tania Tudorache, Abraham Bernstein, Chris Welty, Craig A. Knoblock, Denny Vrandecic, Paul T. Groth, Natasha F. Noy, Krzysztof Janowi...

    2014

  12. [20]

    Pinar Yanardag and S. V. N. Vishwanathan. Deep graph kernels. In DBLP:conf/kdd/2015 , pages 1365--1374. ISBN 978-1-4503-3664-2. doi:10.1145/2783258.2783417. URL http://doi.acm.org/10.1145/2783258.2783417

    2015

  13. [21]

    BioSNAP Datasets : Stanford biomedical network dataset collection

    Marinka Zitnik, Rok Sosi c , Sagar Maheshwari, and Jure Leskovec. BioSNAP Datasets : Stanford biomedical network dataset collection. http://snap.stanford.edu/biodata, August 2018

    2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.