A5025295947- Semantic Web and Ontologies
- Advanced Database Systems and Queries
- Advanced Graph Neural Networks
- Data Quality and Management
- Data Management and Algorithms
- Service-Oriented Architecture and Web Services
- Topic Modeling
- Biomedical Text Mining and Ontologies
- Logic, Reasoning, and Knowledge
- Natural Language Processing Techniques
- Graph Theory and Algorithms
- Web Data Mining and Analysis
- Manufacturing Process and Optimization
- Flexible and Reconfigurable Manufacturing Systems
- Scientific Computing and Data Management
- Big Data and Business Intelligence
- Business Process Modeling and Analysis
- Rough Sets and Fuzzy Logic
- Bayesian Modeling and Causal Inference
- Cloud Computing and Resource Management
- Access Control and Trust
- Image Retrieval and Classification Techniques
- Machine Learning in Materials Science
- Data Visualization and Analytics
- Data Mining Algorithms and Applications
Robert Bosch (Germany)
2019-2025
University of Oslo
2018-2024
Robert Bosch (India)
2020-2024
École Nationale d'Ingénieurs de Tarbes
2022
University of Oxford
2012-2020
Oxford Research Group
2016-2019
Science Oxford
2017
Free University of Bozen-Bolzano
2007-2013
Institut national de recherche en informatique et en automatique
2011-2012
Abstract Digitalisation trends of Industry 4.0 and Internet Things led to an unprecedented growth manufacturing data. This opens new horizons for data-driven methods, such as Machine Learning (ML), in monitoring processes. In this work, we propose ML pipelines quality Resistance Spot Welding. Previous approaches mostly focused on estimating welding based data collected from laboratory or experimental settings. Then, they treated operations independent events while is a continuous process...
Graph self-supervised learning (SSL), including contrastive and generative approaches, offers great potential to address the fundamental challenge of label scarcity in real-world graph data. Among both sets SSL techniques, masked autoencoders (e.g., GraphMAE)—one type methods—have recently produced promising results. The idea behind this is reconstruct node features (or structures)—that are randomly from input—with autoencoder architecture. However, performance feature reconstruction...
Graph Neural Networks (GNNs) have achieved promising performance in various real-world applications. However, recent studies shown that GNNs are vulnerable to adversarial attacks. In this paper, we study a recently-introduced realistic attack scenario on graphs -- graph injection (GIA). the GIA scenario, adversary is not able modify existing link structure and node attributes of input graph, instead performed by injecting nodes into it. We present an analysis topological vulnerability under...
Entity alignment, aiming to identify equivalent entities across different knowledge graphs (KGs), is a fundamental problem for constructing Web-scale KGs. Over the course of its development, label supervision has been considered necessary accurate alignments. Inspired by recent progress self-supervised learning, we explore extent which can get rid entity alignment. Commonly, information (positive pairs) used supervise process pulling aligned in each positive pair closer. However, our...
Graph-based anomaly detection has been widely used for detecting malicious activities in real-world applications. Existing attempts to address this problem have thus far focused on structural feature engineering or learning the binary classification regime. In work, we propose leverage graph contrastive and present supervised GCCAD model contrasting abnormal nodes with normal ones terms of their distances global context (e.g., average all nodes). To handle scenarios scarce labels, further...
An important application of semantic technologies in industry has been the formalisation information models using OWL 2 ontologies and use RDF for storing exchanging data. Moreover, legacy data can be virtualised as following ontology-based ac cess (OBDA) approach. In all these applications, it is to provide domain experts with query formulation tools expressing their needs terms queries over ontologies. this work, we present such a tool, OptiqueVQS, which designed based on our experience...
In recent years, graph neural networks (GNNs) have made great progress in recommendation. The core mechanism of GNNs-based recommender system is to iteratively aggregate neighboring information on the user-item interaction graph. However, existing GNNs treat users and items equally cannot distinguish diverse local patterns each node, which makes them suboptimal recommendation scenario. To resolve this challenge, we present a node-wise adaptive network framework ApeGNN. ApeGNN develops...
Data analytics can be problematic in real-world settings, where data sources are often distributed, heterogeneous, and dynamic. Semantic technologies now offer a practical solution that addresses scalability usability issues, has been successfully deployed industry applications.
An increasing number of applications rely on RDF, OWL 2, and SPARQL for storing querying data. SPARQL, however, is not targeted towards end-users, suitable query interfaces are needed. Faceted search a prominent approach end-user data access, several RDF-based faceted systems have been developed. There is, lack rigorous theoretical underpinning in the context RDF 2. In this paper, we provide such solid foundations. We formalise context, identify fragment first-order logic capturing...
Real-time processing of data coming from multiple heterogeneous streams and static databases is a typical task in many industrial scenarios such as diagnostics large machines. A complex diagnostic may require collection up to hundreds queries over data. Although these retrieve the same kind, temperature measurements, they access structurally different sources. In this work we show how Semantic Technologies implemented our system optique can simplify by providing an abstraction...
In this paper we demonstrate a system SemFacet, that is proof of concept prototype for our semantic faceted search approach. SemFacet implemented on top the Yago knowledge base, powered by OWL 2 RL triple store RDFox, and full text engine Lucene. has provided very encouraging results. Via logical reasoning can automatically (i) extract facets, (ii) update query interface with facets relevant current stage users construction's session. supports queries are much more expressive than ones...
Accessing and utilizing enterprise or Web data that is scattered across multiple sources an important task for both applications users. Ontology-based integration, where ontology mediates between the raw its consumers, a promising approach to facilitate such scenario s. This crucially relies on useful mappings relate data, latter being typically stored in relational databases. A number of systems support construction have recently been developed. generic effective benchmark reliable...