An Ontology-Based Deep Learning Approach for Knowledge Graph Completion with Fresh Entities

Amador-Domínguez, Elvira; Hohenecker, Patrick; Lukasiewicz, Thomas; Manrique, Daniel

doi:10.1007/978-3-030-23887-2_15

Cited by 4 publications

(7 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, ontology describes the basic framework of people's cognition of a domain, and, in contrast, a knowledge graph is rich in entities and relation instances. So far, ontology has been widely used in the research of knowledge graphs [31][32][33][34]. As noted by Paulheim [35], ontology is mainly involved in the construction and data introduction phases of KGs.…”

Section: Classical Research Limitationsmentioning

confidence: 99%

Research on Knowledge Graphs with Concept Lattice Constraints

et al. 2021

View full text Add to dashboard Cite

The application of knowledge graphs has been restricted in some domains, especially the industrial and academic domains. One of the reasons is that they require a high reliability of knowledge, which cannot be satisfied by the existing knowledge graph research. By comparison, traditional knowledge engineering has a high correctness, but low efficiency is an inevitable drawback. Therefore, it is meaningful to organically connect traditional knowledge engineering and knowledge graphs. Therefore, we propose a theory from Attribute Implications to Knowledge Graphs, named AIs-KG, which can construct knowledge graphs based on implications. The theory connects formal concept analysis and knowledge graphs. We firstly analyze the mutual transformation based on the ideas of symmetry with a strict proof among the attribute implication, the formal context and the concept lattice, which forms the closed cycle between the three. Particularly, we propose an Augment algorithm (IFC-A) to generate the Implication Formal Context through the attribute implication, which can make knowledge more complete. Furthermore, we regard ontology as a bridge to realize the transformation from the concept lattice to the knowledge graph through some mapping methods. We conduct our experiments on the attribute implication from the rule base of an animal recognition expert system to prove the feasibility of our algorithms.

show abstract

Section: Classical Research Limitationsmentioning

confidence: 99%

Research on Knowledge Graphs with Concept Lattice Constraints

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Under the matrix factorization framework, the text features of nodes are introduced into network representation learning. The Context-aware Embedding (CANE) model [118] uses the text information of network nodes to interpret the relationship between nodes, and learns context-related network representations for network nodes according to different neighbors. Literature [119] uses ontology information to improve the effect of link classification and link prediction.…”

Section: ) Application Of Semantic Information In the Knowledge Grapmentioning

confidence: 99%

Knowledge Graph Completion: A Review

et al. 2020

View full text Add to dashboard Cite

Knowledge graph completion (KGC) is a hot topic in knowledge graph construction and related applications, which aims to complete the structure of knowledge graph by predicting the missing entities or relationships in knowledge graph and mining unknown facts. Starting from the definition and types of KGC, existing technologies for KGC are analyzed in categories. From the evolving point of view, the KGC technologies could be divided into traditional and representation learning based methods. The former mainly includes rule-based reasoning method, probability graph model, such as Markov logic network, and graph computation based method. The latter further includes translation model based, semantic matching model based, representation learning based and other neural network model based methods. In this paper, different KGC technologies are introduced, including their advantages, disadvantages and applicable fields. Finally the main challenges and problems faced by the KGC are discussed, as well as the potential research directions.

show abstract

“…It has been observed as advantageous to transform categorical variables using suitable feature engineering before applying neural network [35]. For this, we used one-hot encoding, a robust feature engineering scheme, for generating the suitable feature vector indices [16]. These categorical features are mapped into sensor state vector indices representing the concurrent sensor activation patterns for a particular activity.…”

Section: One-hot Code Vectorizationmentioning

confidence: 99%

“…In addition to separation logic, use of a query language, SPARQL also provides support for disengaging these semantics and assertions for interpreting any rule-based complex activities [15]. In work by Amador et al [16], the authors used SPARQL for retrieving class entities and their types, which were later transformed into vector form before using deep learning approaches. Similarly, Socher et al [17] have bridged neural networks with an ontological knowledge-base for the identification of additional facts.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SemImput: Bridging Semantic Imputation with Deep Learning for Complex Human Activity Recognition

Razzaq

Cleland

Nugent

et al. 2020

Sensors

View full text Add to dashboard Cite

The recognition of activities of daily living (ADL) in smart environments is a well-known and an important research area, which presents the real-time state of humans in pervasive computing. The process of recognizing human activities generally involves deploying a set of obtrusive and unobtrusive sensors, pre-processing the raw data, and building classification models using machine learning (ML) algorithms. Integrating data from multiple sensors is a challenging task due to dynamic nature of data sources. This is further complicated due to semantic and syntactic differences in these data sources. These differences become even more complex if the data generated is imperfect, which ultimately has a direct impact on its usefulness in yielding an accurate classifier. In this study, we propose a semantic imputation framework to improve the quality of sensor data using ontology-based semantic similarity learning. This is achieved by identifying semantic correlations among sensor events through SPARQL queries, and by performing a time-series longitudinal imputation. Furthermore, we applied deep learning (DL) based artificial neural network (ANN) on public datasets to demonstrate the applicability and validity of the proposed approach. The results showed a higher accuracy with semantically imputed datasets using ANN. We also presented a detailed comparative analysis, comparing the results with the state-of-the-art from the literature. We found that our semantic imputed datasets improved the classification accuracy with 95.78% as a higher one thus proving the effectiveness and robustness of learned models.

show abstract

An Ontology-Based Deep Learning Approach for Knowledge Graph Completion with Fresh Entities

Cited by 4 publications

References 5 publications

Research on Knowledge Graphs with Concept Lattice Constraints

Research on Knowledge Graphs with Concept Lattice Constraints

Knowledge Graph Completion: A Review

SemImput: Bridging Semantic Imputation with Deep Learning for Complex Human Activity Recognition

Contact Info

Product

Resources

About