GEDLIB: A C++ Library for Graph Edit Distance Computation

Blumenthal, David B.; Bougleux, Sébastien; Gamper, Johann; Brun, Luc

doi:10.1007/978-3-030-20081-7_2

Cited by 12 publications

(13 citation statements)

References 25 publications

(15 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…-Meta-parameters for SUBGRAPH and WALKS: As suggested in [21] and [32], for each dataset, we determined the parameters K of SUBGRAPH and WALKS as the K ∈ [5] that yielded the tightest average upper bounds on a set of training graphs. To cope with SUBGRAPH's exponential runtime complexity, we set a time limit of 1 ms for the computation of each cell of its LSAPE instance C. -Options and meta-parameters for RING: As highlighted in [4,6], RING performs best if the node and edge set distances are computed via optimal LSAPE solvers or multiset intersection based proxies.…”

Section: Choice Of Options and Parametersmentioning

confidence: 99%

See 1 more Smart Citation

Comparing heuristics for graph edit distance computation

Blumenthal¹,

Boria²,

Gamper³

et al. 2019

The VLDB Journal

Self Cite

View full text Add to dashboard Cite

Because of its flexibility, intuitiveness, and expressivity, the graph edit distance (GED) is one of the most widely used distance measures for labeled graphs. Since exactly computing GED is NP-hard, over the past years, various heuristics have been proposed. They use techniques such as transformations to the linear sum assignment problem with error-correction, local search, and linear programming to approximate GED via upper or lower bounds. In this paper, we provide a systematic overview of the most important heuristics. Moreover, we empirically evaluate all compared heuristics within an integrated implementation.

show abstract

Section: Choice Of Options and Parametersmentioning

confidence: 99%

“…https://github.com/dbblumenthal/gedlib/ [5]. Tests were run on a machine with two Intel Xeon E5-2667 v3 processors with 8 cores each and 98 GB of main memory running GNU/Linux.…”

Section: Implementation and Hardware Specificationsmentioning

confidence: 99%

Comparing heuristics for graph edit distance computation

Blumenthal¹,

Boria²,

Gamper³

et al. 2019

The VLDB Journal

Self Cite

View full text Add to dashboard Cite

show abstract

“…We conduct all experiments in this paper using the PHOCNet implementation 3 by Sudholt and Fink [24]. For the sample selection, we use the IPM implementation 4 by Zaeemzadeh et al [29] and we compute the HED between graphs using the HED implementation in GEDLIB 5 by Blumenthal et al [4,5], which we modified to support the cost model by Stauffer et al [23]. The code used to conduct all experiments as well as the raw data is available online 6 .…”

Section: Experiments Setupmentioning

confidence: 99%

Representative Image Selection for Data Efficient Word Spotting

Westphal

Grahn

Lavesson

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

This paper compares three different word image representations as base for label free sample selection for word spotting in historical handwritten documents. These representations are a temporal pyramid representation based on pixel counts, a graph based representation, and a pyramidal histogram of characters (PHOC) representation predicted by a PHOCNet trained on synthetic data. We show that the PHOC representation can help to reduce the amount of required training samples by up to 69% depending on the dataset, if it is learned iteratively in an active learning like fashion. While this works for larger datasets containing about 1 700 images, for smaller datasets with 100 images, we find that the temporal pyramid and the graph representation perform better.

show abstract

“…A major difficulty, which is not straightforward from (5), is that ω and c are interdependent. For two different edit cost vectors, respective optimal ω may not be equivalent since the costs influence the presence or absence of each edit operation.…”

mentioning

confidence: 99%

“…with d GED depending on c and ω as given in (5), where ω exists for each pair of graphs G i and G j in G N , which will be denoted as ω(i, j). Moreover, to ensure that the minimum cost edit transformation π in (1) can be found, all edit costs need to be positive, and substituting an element should not be more expensive than removing and inserting it [21].…”

mentioning

confidence: 99%

A Graph Pre-image Method Based on Graph Edit Distances

Jia

Gaüzère

Honeiné

2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The pre-image problem for graphs is increasingly attracting attention owing to many promising applications. However, it is a challenging problem due to the complexity of graph structure. In this paper, we propose a novel method to construct graph pre-images as median graphs, by aligning graph edit distances (GEDs) in the graph space with distances in the graph kernel space. The two metrics are aligned by optimizing the edit costs of GEDs according to the distances between the graphs within the space associated with a particular graph kernel. Then, the graph pre-image can be estimated using a median graph method founded on the GED. In particular, a recently introduced method to compute generalized median graphs with iterative alternate minimizations is revisited for this purpose. Conducted experiments show very promising results while opening the computation of graph pre-image to any graph kernel and to graphs with non-symbolic attributes.

show abstract

GEDLIB: A C++ Library for Graph Edit Distance Computation

Cited by 12 publications

References 25 publications

Comparing heuristics for graph edit distance computation

Comparing heuristics for graph edit distance computation

Representative Image Selection for Data Efficient Word Spotting

A Graph Pre-image Method Based on Graph Edit Distances

Contact Info

Product

Resources

About