Semantic Mapping Relational to a Directed Property Hypergraph Model

Wardani, Dewi Kusuma; Küng, Josef

doi:10.1109/cit/iucc/dasc/picom.2015.24

Cited by 6 publications

(6 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although a graph model basically has no schema, but the mapping process should be started from the schema, and the rest populated data will be following the schema. In summary, up to our knowledge, the semantic mapping which has been proposed [10] is the first work which maps with considering the higher abstraction of the relational model and has shown good performances comparing the naïve model. Showing good performances is not enough to acknowledge the model keeps up the semantic of the real world.…”

Section: Introductionmentioning

confidence: 90%

“…In the real world is also very common that a relationship relates to more than two things. In this part we summarize some definitions of our previous work [10] as below: DEFINITION 1 (The Relational Model). Let S (R1, R2 … Ri) be a relational model which consists of a set of Relations, i is the degree of relation, a set of Primary Key PK (PK1, PK2 … PKiPK), a set of foreign key FK (FK1, FK2 … FkiFK) and others sets of integrity constrains (IC).…”

Section: Link Structure In Relational Modelmentioning

confidence: 99%

“…From the real IMDB data set, we learn that in the complex and huge relational schema there is a bigger possibility many relations which should be mapped as a link, not as a node. Even though the data set of the experiment are the result of this approach [10], but basically all obtained graph models from the other approaches can be evaluated as well.…”

Section: B the Social Databasementioning

confidence: 99%

“…The remaining sections are organized as follows: section two explains the related work. Section three summarizes the mapping approach from the previous work [10]. Section four describes the evaluating method to examine whether the graph keeps up the semantic of the real world.…”

Section: Introductionmentioning

confidence: 99%

“…The example mapping and converting (a) relational model to (b). a property graph model based on the approach of the previous work[10] …”

mentioning

confidence: 99%

See 4 more Smart Citations

Evaluating The Semantic Mapping

Wardani¹

2018

EECSI

View full text Add to dashboard Cite

The increasing of the importance of links in the network of data influences the idea that links should be considered more in the mapping relational to graph model. Semantic abstraction gaps often occur during the mapping process where the link in the real world is mapped as a node in a graph model. This paper focused on evaluating the result of mapping and converting without losing the semantics. We propose the evaluation of our approach by using schema.org as the semantic standard. The experiments in three data sets show that the semantic mapping approach is pretty effective. We obtain quite good score matching without considering the gap index (the average is 0.6922) and with considering the gap index (the average is 0.5264) and the average precision score, 0.7042, is pretty good too.

show abstract

Section: Introductionmentioning

confidence: 90%

Section: Link Structure In Relational Modelmentioning

confidence: 99%

Section: B the Social Databasementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…The example mapping and converting (a) relational model to (b). a property graph model based on the approach of the previous work[10] …”

mentioning

confidence: 99%

See 3 more Smart Citations

Evaluating The Semantic Mapping

Wardani¹

2018

EECSI

View full text Add to dashboard Cite

show abstract

TransH-RA: A Learning Model of Knowledge Representation by Hyperplane Projection and Relational Attributes

et al. 2023

View full text Add to dashboard Cite

The TransE model plays a key role in dealing with data sparsity and promotes the development of knowledge graphs completion. However, TransE has some difficulties in dealing with oneto-many, many-to-one, many-to-many and transmission relationships. In order to solve this problem, this paper proposes a knowledge representation learning model based on hyperplane projection and relational attributes, namely TransH-RA. First of all, we introduce the idea of hyperplane projection based on the TransE model, this idea is inspired by TransH, which makes different entities have different roles in a specific relationship, thus reducing the constraints of TransE translation rules, and map the head entity h and tail entity t to the plane of special relation r; Secondly, considering that it is not easy to identify different similar entities, the neighborhood information of entities is added to learn the neighborhood of entities around different entities; Then, in order to further strengthen the ability to deal with complex relationships, attribute features of relationships are added and attribute knowledge is embedded; Eventually, during the training of the model, the probability method is chosen to replace the head and tail entities. Link prediction experiments are conducted on the public datasets FB15K and WN18, and the triple classification experiments on the datasets WN11, FB13 and FB15K are carried out to analyze and verify the effectiveness of the proposed method. The evaluation results show that our method achieves the better results on MeanRank, Hits@10 and ACC indicators compared with TransE and TransH.

show abstract