Optimizing Similarity Computations for Ontology Matching - Experiences from GOMMA

Hartung, Michael; Kolb, Lars; Groß, Anika; Rahm, Erhard

doi:10.1007/978-3-642-39437-9_7

Cited by 7 publications

(3 citation statements)

References 14 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because ontologies can contain millions of entities, it is often infeasible to compare every entity in one ontology to every entity in the other. Therefore, alignment systems sometimes employ a filtering or hashing step to determine which entities to compare [Duan et al 2012;Hartung et al 2013].…”

Section: Ontology Alignmentmentioning

confidence: 99%

The properties of property alignment on the semantic web

Cheatham

2018

IJMSO

View full text Add to dashboard Cite

2014. The Properties of Property Alignment on the Semantic Web. Ontology alignment is an important step in enabling computers to query and reason across the many linked datasets on the semantic web. This is a difficult challenge because the ontologies underlying different linked datasets can vary in terms of subject area coverage, level of abstraction, ontology modeling philosophy, and even language. The alignment approach presented here centers on string similarity metrics. Nearly all ontology alignment systems use a string similarity metric in one form or another, but it seems that the choice of a particular metric is often arbitrary. We begin this dissertation with the most comprehensive survey to date on the performance of string similarity metrics and string preprocessing strategies for ontology alignment. Based on this work we present practical guidelines for choosing string metrics in the face of different types of ontologies and different alignment goals. Additionally, we show that string similarity metrics alone can perform competitively with state-of-the-art alignment systems on the most popular benchmarks in the field. One of the contributions of our string similarity metric survey is quantification of the difference in performance between aligning classes and aligning properties (relations between classes). Put simply: aligning properties is hard, and existing string similarity metrics are not of great help. We therefore take on the task of developing a new string-based alignment approach that performs better on properties. Unfortunately, evaluating that approach is difficult because the only existing alignment benchmark that includes properties is, in our view, unrealistic since all relations in the reference alignment are presented as completely certain. Human experts do not have this degree of confidence when asked to align an ontology. We therefore present a more nuanced version of this benchmark that we have created through a combination of expert survey and crowdsourcing. We then present our new string-based property alignment system and evaluate its performance on both the current benchmark and our proposed revision. Our property-centric string metric can be configured for either high precision or high recall. The results show a five-fold increase in precision and a doubling of recall over an approach based on the best current string metric. Finally, we apply our system to a real-world test case and analyze the results.

show abstract

Section: Ontology Alignmentmentioning

confidence: 99%

The properties of property alignment on the semantic web

Cheatham

2018

IJMSO

View full text Add to dashboard Cite

show abstract

“…Furthermore, computation of a distance is also performed in parallel with the use of local memory within a GPU. Combined CPU/GPU parallel similarity computations are also discussed in paper [12] in the context of ontology matching. The authors consider and address issues such as limited data types on the GPU side as well as limited memory.…”

Section: Related Workmentioning

confidence: 99%

Parallelization of large vector similarity computations in a hybrid CPU+GPU environment

Czarnul

2017

J Supercomput

View full text Add to dashboard Cite

The paper presents design, implementation and tuning of a hybrid parallel OpenMP+CUDA code for computation of similarity between pairs of a large number of multidimensional vectors. The problem has a wide range of applications, and consequently its optimization is of high importance, especially on currently widespread hybrid CPU+GPU systems targeted in the paper. The following are presented and tested for computation of all vector pairs: tuning of a GPU kernel with consideration of memory coalescing and using shared memory, minimization of GPU memory allocation costs, optimization of CPU-GPU communication in terms of size of data sent, overlapping CPU-GPU communication and kernel execution, concurrent kernel execution, determination of best sizes for data batches processed on CPUs and GPUs along with best GPU grid sizes. It is shown that all codes scale in hybrid environments with various relative performances of compute devices, even for a case when comparisons of various vector pairs take various amounts of time. Tests were performed on two high-performance hybrid systems with: 2 x Intel Xeon E5-2640 CPU + 2 x NVIDIA Tesla K20m and latest generation 2 x Intel Xeon CPU E5-2620 v4 + NVIDIA's Pascal generation GTX 1070 cards. Results demonstrate expected improvements and beneficial optimizations important for users incorporating such types of computations into their parallel codes run on similar systems.

show abstract

“…GKBs are mainly represented in three different forms, namely geographic semantic webs, geographic ontologies, and digital gazetteers. Geographic semantic webs (Table 1), which mainly include GeoNames (http://www.geonames.org/), LinkedGeoData management (GOMMA) [23]. In GIScience, there have also been a great number of studies on this subject.…”

Section: Introductionmentioning

confidence: 99%

Progress and Challenges on Entity Alignment of Geographic Knowledge Bases

Sun

Zhu

Song

2019

IJGI

View full text Add to dashboard Cite

Geographic knowledge bases (GKBs) with multiple sources and forms are of obvious heterogeneity, which hinders the integration of geographic knowledge. Entity alignment provides an effective way to find correspondences of entities by measuring the multidimensional similarity between entities from different GKBs, thereby overcoming the semantic gap. Thus, many efforts have been made in this field. This paper initially proposes basic definitions and a general framework for the entity alignment of GKBs. Specifically, the state-of-the-art of algorithms of entity alignment of GKBs is reviewed from the three aspects of similarity metrics, similarity combination, and alignment judgement; the evaluation procedure of alignment results is also summarized. On this basis, eight challenges for future studies are identified. There is a lack of methods to assess the qualities of GKBs. The alignment process should be improved by determining the best composition of heterogeneous features, optimizing alignment algorithms, and incorporating background knowledge. Furthermore, a unified infrastructure, techniques for aligning large-scale GKBs, and deep learning-based alignment techniques should be developed. Meanwhile, the generation of benchmark datasets for the entity alignment of GKBs and the applications of this field need to be investigated. The progress of this field will be accelerated by addressing these challenges.

show abstract

Optimizing Similarity Computations for Ontology Matching - Experiences from GOMMA

Cited by 7 publications

References 14 publications

The properties of property alignment on the semantic web

The properties of property alignment on the semantic web

Parallelization of large vector similarity computations in a hybrid CPU+GPU environment

Progress and Challenges on Entity Alignment of Geographic Knowledge Bases

Contact Info

Product

Resources

About