| |
|
|
|
|
|
|
|
|
1. |
Record Nr. |
UNINA9910585783003321 |
|
|
Titolo |
Database and expert systems applications . Part I : 33rd International Conference, DEXA 2022, Vienna, Austria, August 22-24, 2022, proceedings / / Christine Strauss [and four others], editors |
|
|
|
|
|
|
|
Pubbl/distr/stampa |
|
|
Cham, Switzerland : , : Springer International Publishing, , [2022] |
|
©2022 |
|
|
|
|
|
|
|
|
|
ISBN |
|
|
|
|
|
|
Descrizione fisica |
|
1 online resource (469 pages) |
|
|
|
|
|
|
Collana |
|
Lecture Notes in Computer Science ; ; v.13426 |
|
|
|
|
|
|
Disciplina |
|
|
|
|
|
|
Soggetti |
|
Database management |
Expert systems (Computer science) |
|
|
|
|
|
|
|
|
Lingua di pubblicazione |
|
|
|
|
|
|
Formato |
Materiale a stampa |
|
|
|
|
|
Livello bibliografico |
Monografia |
|
|
|
|
|
Nota di contenuto |
|
Intro -- Preface -- Organization -- Abstracts of Keynote Talks -- Responsible AI -- Following the Rules: From Policies to Norms -- Contents - Part I -- Contents - Part II -- Knowledge Graphs -- Jointly Learning Propagating Features on the Knowledge Graph for Movie Recommendation -- 1 Introduction -- 2 Methodology -- 2.1 Problem Formulation -- 2.2 Heterogeneous Graph Construction -- 2.3 The Proposed Framework -- 3 Experiments -- 3.1 Datasets -- 3.2 Experimental Settings -- 3.3 Experimental Results -- 3.4 Ablation Study -- 4 Conclusion -- References -- Syntax-Informed Question Answering with Heterogeneous Graph Transformer -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Graph Creation Module -- 3.2 Graph Processing and Answer Prediction -- 4 Experiments and Discussions -- 4.1 Setup -- 4.2 Evaluation -- 4.3 LingHGT, SemHGT, PragHGT: Towards Linguistics-Informed Language Models -- 5 Conclusion -- References -- Automated Process Knowledge Graph Construction from BPMN Models -- 1 Introduction -- 2 Background -- 3 Industrial Use Case -- 4 BPMN2KG Tool -- 4.1 Requirements -- 4.2 Process Knowledge Graph Construction -- 4.3 Implementation -- 5 Application Scenarios -- 6 Related Work -- 7 Conclusions -- A SPARQL Queries and Results -- References -- CAKE: A Context-Aware Knowledge Embedding Model of Knowledge Graph -- 1 Introduction -- |
|
|
|
|
|
|
|
|
|
2 Related Work -- 3 Context-Aware Knowledge Embedding (CAKE) -- 3.1 LDA-Based Context Learning -- 3.2 HDP-Based Context Learning -- 3.3 Context-Aware Knowledge Embedding -- 4 Experimental Evaluation -- 4.1 Experiment Protocol -- 4.2 Experiment Settings -- 4.3 Experimental Results -- 5 Conclusion -- A Appendix: Optimization of CAKE -- References -- The Digitalization of Bioassays in the Open Research Knowledge Graph -- 1 Introduction -- 2 Bioassay Digitalization in the ORKG -- 3 Conclusion -- References. |
Privacy-Preservation Approaches -- Privacy Issues in Smart Grid Data: From Energy Disaggregation to Disclosure Risk -- 1 Introduction -- 2 Related Work -- 2.1 Non-intrusive Load Monitoring -- 2.2 Privacy Preserving Data Publishing in Smart Grid -- 3 Proposed Method for Disclosure Risk Assessment -- 3.1 Appliance Selection -- 3.2 Seq2Seq Disaggregation Algorithm -- 3.3 Event Detection -- 3.4 Middle-Point Thresholding -- 3.5 Variance-Sensitive Thresholding -- 3.6 Activation Time Extraction -- 3.7 Disaggregation Risk -- 4 Experimental Setup -- 4.1 Datasets -- 4.2 Training and Testing Period -- 4.3 Threshold Computation -- 5 Results and Discussion -- 5.1 Inference Attack on the Same Building -- 5.2 Inference Attack on Different Buildings in the Same Dataset -- 5.3 Inference Attack on Different Buildings from Different Datasets -- 6 Conclusion -- References -- CoK: A Survey of Privacy Challenges in Relation to Data Meshes -- 1 Introduction -- 2 State-of-the-Art -- 2.1 Data Mesh Databases -- 2.2 Privacy in Mesh Networking -- 2.3 Syntactic Data Anonymisation -- 2.4 Semantic Data Anonymisation and Differential Privacy -- 2.5 Unique Column Combinations -- 2.6 High-Dimensional Data -- 2.7 Quasi-Identifier Discovery -- 3 Data Meshes -- 4 Experiments -- 5 Conclusion and Future Directions -- References -- Why- and How-Provenance in Distributed Environments -- 1 Introduction -- 2 Background and Related Work -- 2.1 Data Provenance -- 2.2 Distributed Databases -- 2.3 Related Work -- 3 Provenance in Distributed Databases -- 3.1 Architecture -- 3.2 Annotations -- 3.3 Build Provenance Information -- 4 Experimental Evaluation -- 5 Conclusions and Future Work -- References -- Provenance-Based SPARQL Query Formulation -- 1 Introduction -- 2 Model -- 3 Structural Edits -- 4 Grounding Generator -- 5 Procuring Feedback -- 6 Experiments -- 7 Related Work -- 8 Conclusion. |
References -- Anonymisation of Heterogeneous Graphs with Multiple Edge Types -- 1 Introduction -- 2 K-RDF-Neighbourhood Anonymisation with Multiple Edge Types -- 3 Conclusions -- References -- Deep Learning -- A Divergent Index Advisor Using Deep Reinforcement Learning -- 1 Introduction -- 2 Related Work -- 3 DRL Divergent Index Selection Framework -- 3.1 Pre-processing Module -- 3.2 DRL Divergent Index Selection Module -- 3.3 The Cost Estimation Module -- 4 Experimental Results -- 4.1 Experimental Setup -- 5 Conclusion and Future Research -- References -- Deep Active Learning Framework for Crowdsourcing-Enhanced Image Classification and Segmentation -- 1 Introduction -- 2 Related Work -- 3 Proposed Method -- 3.1 Notation -- 3.2 Neural Networks for Image Tasks -- 3.3 Active Learning -- 4 Performance Evaluation -- 4.1 Crowdsourcing Image Classification via Active Learning -- 4.2 Crowdsourcing Image Segmentation via Active Learning -- 4.3 Analysis and Discussion -- 5 Conclusions and Perspectives -- References -- Sentiment and Knowledge Based Algorithmic Trading with Deep Reinforcement Learning -- 1 Introduction -- 2 Related Work -- 3 Proposed Approach -- 3.1 Reinforcement Learning -- 3.2 Sentiment Analysis -- 3.3 Knowledge Graphs -- 4 Empirical Evaluation -- 4.1 Data -- 4.2 MDP Formulation -- 5 Results and Analysis -- 5.1 Training Data Analysis -- |
|
|
|
|
|
|
|
5.2 Test Data Analysis -- 5.3 Sharpe Ratio -- 6 Discussion and Conclusion -- References -- DeepCore: A Comprehensive Library for Coreset Selection in Deep Learning -- 1 Introduction -- 2 Review of Coreset Selection Methods -- 2.1 Problem Statement -- 2.2 Survey: Methodologies -- 2.3 Survey: Applications -- 3 DeepCore Library -- 4 Experiment Results -- 4.1 CIFAR10 Results -- 4.2 ImageNet Results -- 4.3 Cross-architecture Generalization -- 4.4 Sensitiveness to Pre-trained Models -- 5 Extended Related Work. |
6 Conclusion -- References -- Context Iterative Learning for Aspect-Level Sentiment Classification -- 1 Introduction -- 2 Related Work -- 3 Context Attention Modules (CAM) -- 3.1 Intra-Multi-Headed Attention Mechanism (Intra-MHA) and Inter-Multi-Headed Attention Mechanism (Inter-MHA) -- 3.2 Context Features Dynamic Mask/Context Features Dynamic Weighted (CDM/CDW) -- 3.3 Position-Wise Feed-Forward Networks (PFFN) and Aspect-Context Representation Output -- 4 Context Iterative Learning Network (CILN) -- 4.1 Pooling Layer and Training -- 5 Experiment -- 5.1 Datasets and Experimental Settings -- 5.2 Baseline and Result -- 6 Conclusion -- References -- Smart Cities and Human Computing -- EcoLight: Eco-friendly Traffic Signal Control Driven by Urban Noise Prediction -- 1 Introduction -- 2 Related Work -- 2.1 Noise Prediction -- 2.2 Traffic Signal Control -- 3 Formalization of the Problem -- 4 EcoLight Approach -- 4.1 Traffic Noise Prediction -- 4.2 Traffic Signal Control -- 5 Experimental Evaluation -- 5.1 Dataset -- 5.2 Experimental Setups -- 5.3 Baseline Methods for Comparison -- 5.4 Evaluation -- 5.5 Results and Discussion -- 6 Conclusion -- References -- Mining Fluctuation Propagation Graph Among Time Series with Active Learning -- 1 Introduction -- 2 Related Work -- 3 Empirical Study of Mining Methods -- 3.1 Experimental Setup -- 3.2 Results -- 4 FPG-Miner: Mine with Active Learning -- 4.1 Recommendation Framework -- 4.2 Continuous Association Rule Classifier -- 5 Experiment -- 5.1 Experimental Setup -- 5.2 Results -- 5.3 Case Study: Root Cause Analysis -- 6 Conclusion -- References -- Towards Efficient Human Action Retrieval Based on Triplet-Loss Metric Learning -- 1 Introduction -- 2 Action Retrieval -- 2.1 Problem Definition -- 2.2 Retrieval Process -- 3 Learning Action Features -- 3.1 Triplet-Loss Learning -- 3.2 Triplet Generation Strategies. |
4 Experimental Evaluation -- 4.1 Dataset -- 4.2 Methodology -- 4.3 Experimental Results -- 4.4 State-of-the-Art Comparison -- 5 Conclusions -- References -- KAPP: Knowledge-Aware Hierarchical Attention Network for Popularity Prediction -- 1 Introduction -- 2 Methodology -- 2.1 Knowledge-Aware User Embedding -- 2.2 Attention-Based Retweet Path Encoding -- 2.3 Prediction with Point Process -- 3 Experiments -- 3.1 Experiment Setup -- 3.2 Numerical Results -- 4 Conclusion -- References -- Advanced Machine Learning -- A Heterogeneous Network Representation Learning Approach for Academic Behavior Prediction -- 1 Introduction -- 2 Related Work -- 3 The Mechanism of HNEABP Approach -- 3.1 Balanced Walk Method Based on Edge Number -- 3.2 Balanced Walk Method Based on Edge Loss -- 3.3 Node Pair and Semantic Information Learning Based on KGE -- 3.4 Loss Function -- 3.5 HNEABP Algorithm -- 4 Verification Experiment on Node Representation Learning -- 4.1 Datasets and Baseline Methods -- 4.2 Experimental Setup and Evaluation Criteria -- 4.3 Analysis of Experimental Results -- 5 Conclusion -- References -- A Market Segmentation Aware Retail Itemset Placement Framework -- 1 Introduction -- 2 Related Work -- 3 Proposed Framework of Itemset Placement Problem -- 4 MATRIX and MIPS -- 5 Performance Evaluation -- 6 Conclusion -- References -- Label Selection Algorithm Based on Iteration Column Subset Selection for Multi-label |
|
|
|
|
|
|
|
|
Classification -- 1 Introduction -- 2 Multi-label Classification Using Iteration Column Subset Selection -- 2.1 Notation -- 2.2 Iteration Column Subset Selection -- 2.3 Building a Recovery Matrix -- 2.4 Learning the Model and Prediction -- 3 Experiments -- 3.1 Six Benchmark Data Sets and Two Evaluation Metrics -- 3.2 Experimental Settings -- 3.3 Results -- 4 Conclusions -- References. |
Accurately Predicting User Registration in Highly Unbalanced Real-World Datasets from Online News Portals. |
|
|
|
|
|
| |