Answers Partitioning and Lazy Joins for Efficient Query Relaxation and Application to Similarity Search

Ferré, Sébastien

doi:10.1007/978-3-319-93417-4_14

Cited by 10 publications

(11 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we shortly recall the theoretical definitions underlying Concepts of Nearest Neighbours (C-NN), as well as the algorithmic and practical aspects of their approximate computation under a given timeout. Further details are available in [7,8]. In the following definitions, we assume a knowledge graph K = E, R, T .…”

Section: Concepts Of Nearest Neighbours (C-nn)mentioning

confidence: 99%

“…We here sketch the algorithmic and practical aspects of computing the set C-NN(e, K) of concepts of nearest neighbours of query entity e in a knowledge graph K. More details are available in [8]. The naive approach would be to compute |E| conceptual distances between e and the other entities.…”

Section: Algorithmic and Practical Aspectsmentioning

confidence: 99%

“…In this paper, we report on extensive experimental results about a novel approach to knowledge graph completion that is based on Concepts of Nearest Neighbours (C-NN), which were introduced in [7], and applied to query relaxation in [8]. This paper is an extended version of [9] that reports on first results about the application of C-NNs to link prediction.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Application of concepts of neighbours to knowledge graph completion1

Ferré

2021

Self Cite

View full text Add to dashboard Cite

The open nature of Knowledge Graphs (KG) often implies that they are incomplete. Knowledge graph completion (a.k.a. link prediction) consists in inferring new relationships between the entities of a KG based on existing relationships. Most existing approaches rely on the learning of latent feature vectors for the encoding of entities and relations. In general however, latent features cannot be easily interpreted. Rule-based approaches offer interpretability but a distinct ruleset must be learned for each relation. In both latent- and rule-based approaches, the training phase has to be run again when the KG is updated. We propose a new approach that does not need a training phase, and that can provide interpretable explanations for each inference. It relies on the computation of Concepts of Nearest Neighbours (C-NN) to identify clusters of similar entities based on common graph patterns. Different rules are then derived from those graph patterns, and combined to predict new relationships. We evaluate our approach on standard benchmarks for link prediction, where it gets competitive performance compared to existing approaches.

show abstract

Section: Concepts Of Nearest Neighbours (C-nn)mentioning

confidence: 99%

Section: Algorithmic and Practical Aspectsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Application of concepts of neighbours to knowledge graph completion1

Ferré

2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ferr´e, [21] developed two algorithms namely answer portioning and lazy join and this method are process to increase the efficiency. This technique scales much better with the size of the query.…”

Section: Related Workmentioning

confidence: 99%

“…3. The answers partitioning and lazy join are the two algorithms proposed in the research [21]. These methods are executed with the seven different queries and the execution time is calculated.…”

Section: Fig 3 the Computation Time Of The Variousmentioning

confidence: 99%

Teaching Learning Based Optimization in Semantic Web of Distributed RDF

Shailaja¹,

Kumar²,

Bhavani³

2019

IJRTE

View full text Add to dashboard Cite

 Abstract: Semantic web data use as a unified data model in various areas, such as Bioinformatics, media data, Wikipedia, social networks, and government open data. Sharing information among people using semantic web helps to understand and manipulation of information. In the semantic web, the Resource Description Framework (RDF) denotes the linked data. The logical data is represented as RDF model to manage the unformatted data and it provides an ability to machine interpretability of data. The major problem on the web is to handle the large volume of the data that also has other challenges like query processing and optimization over widely distributed RDF data. In this research, the Teacher Learning based Optimization (TLBO) algorithm is proposed for the query optimization to reduce query cost, and optimize the computation time of the query. The TLBO technique select the suitable location and size of the population based on the data that effectively provide the solution for the distributed data i.e.., triple pattern of semantic web. The experimental result showed that the TLBO in query optimization performed well in the manner of query computation time compared to existing methods like MARVEL. Additionally, the results showed that the proposed TLBO model achieved nearly 4.93 seconds for executing the multiple queries in LUBM dataset.

show abstract

Link Prediction in Knowledge Graphs with Concepts of Nearest Neighbours

Ferré

2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Answers Partitioning and Lazy Joins for Efficient Query Relaxation and Application to Similarity Search

Cited by 10 publications

References 18 publications

Application of concepts of neighbours to knowledge graph completion1

Application of concepts of neighbours to knowledge graph completion1

Teaching Learning Based Optimization in Semantic Web of Distributed RDF

Link Prediction in Knowledge Graphs with Concepts of Nearest Neighbours

Contact Info

Product

Resources

About