Big data analytics and knowledge discovery : 23rd international conference, DaWaK 2021, virtual event, September 27-30, 2021, proceedings / / Matteo Golfarelli [and three others], editors

• Pubblicazione: Cham, Switzerland : , : Springer, , [2021]; ©2021
• Descrizione fisica: 1 online resource (283 pages)
• Disciplina: 005.74
• Soggetto topico: Database management; Artificial intelligence; Data mining
• Persona (resp. second.): GolfarelliMatteo
• Nota di contenuto: Intro -- Preface -- Organization -- Contents -- Performance -- Bounding Box Representation of Co-location Instances for L Induced Distance Measure -- 1 Introduction -- 2 Related Work -- 3 Definitions -- 3.1 Common Definitions -- 3.2 Co-location Instance Representation -- 4 Instance Identification -- 5 Experiments -- 6 Summary -- References -- Benchmarking Data Lakes Featuring Structured and Unstructured Data with DLBench -- 1 Introduction -- 2 Related Works -- 2.1 Big Data Benchmarks -- 2.2 Textual Benchmarks -- 2.3 Discussion -- 3 Specification of DLBench -- 3.1 Data Model -- 3.2 Workload Model -- 3.3 Performance Metrics -- 3.4 Assessment Protocol -- 4 Proof of Concept -- 4.1 Overview of AUDAL -- 4.2 Setup and Results -- 5 Conclusion -- References -- Towards an Adaptive Multidimensional Partitioning for Accelerating Spark SQL -- 1 Introduction -- 2 Literature Review -- 3 Definitions and Problem Formulation -- 3.1 Preliminaries -- 3.2 Problem Statement -- 4 System Architecture -- 4.1 Partitioning Schema Selection -- 4.2 Controller -- 4.3 Query Rewritten -- 5 Experimental Results -- 6 Conclusions -- References -- Selecting Subexpressions to Materialize for Dynamic Large-Scale Workloads -- 1 Introduction -- 2 Related Work -- 3 Background -- 4 Proactive Selection of Common Subexpression to Materialize -- 4.1 The Online Phase -- 5 Experimental Study -- 6 Conclusion -- References -- Prediction Techniques -- A Chain Composite Item Recommender for Lifelong Pathways -- 1 Introduction -- 2 Formal Definitions -- 2.1 Model -- 2.2 Problem Formulation and Complexity -- 3 Problem Resolution -- 3.1 Computing Action Profiles (P1) -- 3.2 Computing Beneficiary Profile (P2) -- 3.3 Pathway Recommendation (P3) -- 4 Tests -- 4.1 Effectiveness of PRS -- 4.2 Scaling to Larger Instances -- 5 Related Work -- 6 Conclusion -- References. ; Health Analytics on COVID-19 Data with Few-Shot Learning -- 1 Introduction -- 2 Background and Related Works -- 3 Our Health Analytics System -- 3.1 A Serial Prediction Model Based on Samples from the Dataset -- 3.2 A Simultaneous Prediction Model Based on Samples from the Dataset -- 3.3 Prediction Models Based on Samples from Focused Portions of the Dataset -- 4 Evaluation -- 5 Conclusions -- References -- Cognitive Visual Commonsense Reasoning Using Dynamic Working Memory -- 1 Introduction -- 2 Related Work -- 3 Notations and Problem Formulation -- 4 Proposed Framework -- 4.1 Feature Representation Layer -- 4.2 Multimodal Feature Fusion Layer -- 4.3 Encoder Layer -- 4.4 Prediction Layer -- 5 Experimental Results -- 5.1 Experimental Settings -- 5.2 Analysis of Experimental Results -- 5.3 Qualitative Results -- 6 Conclusion -- References -- Knowledge Representation -- Universal Storage Adaption for Distributed RDF-Triple Stores -- 1 Introduction -- 2 Background -- 3 Related Work -- 4 Our Flexible Universal Cost Model -- 5 Workload Analysis -- 5.1 Heat Queries -- 5.2 Heat Query Generation -- 6 Elements of Adaption -- 6.1 Indexes -- 6.2 Join Cache -- 6.3 Replication -- 7 Universal Adaption -- 8 Evaluation -- 9 Conclusion and Future Work -- References -- RDF Data Management is an Analytical Market, not a Transaction One -- 1 Introduction -- 2 Arguments for OLAP RDF Stores -- 2.1 Why RDF Stores Do Need ACID Transactions? -- 2.2 RDF Stores Should Support Analytical Operations -- 3 The Road to Analytics in RDF Stores -- 3.1 Kind of Analytical Operations -- 3.2 Emerging Systems -- 4 Conclusion -- References -- Document Ranking for Curated Document Databases Using BERT and Knowledge Graph Embeddings: Introducing GRAB-Rank -- 1 Introduction -- 2 Literature Review -- 2.1 Traditional Document Ranking Models -- 2.2 Semantic Document Ranking Models. ; 2.3 Knowledge Graph Document Ranking Models -- 3 Problem Definition -- 4 BERT and Knowledge Graph Embeddings Based Document Ranking -- 4.1 Knowledge Graph Construction -- 4.2 Knowledge Graph Concept Embeddings Using a Random Walk -- 4.3 BERT Contextualised Embeddings -- 5 Evaluation -- 6 Future Work and Conclusion -- References -- Advanced Analytics -- Contextual and Behavior Factors Extraction from Pedestrian Encounter Scenes Using Deep Language Models -- 1 Introduction -- 2 Methodology -- 3 Experimental Setting and Results -- 3.1 Dataset and Hyperparameters -- 3.2 Results -- 4 Conclusions and Future Work -- References -- Spark Based Text Clustering Method Using Hashing -- 1 Introduction -- 2 The Hashing Trick -- 3 Spark Based Text Clustering Method for Large Textual Data -- 3.1 Document Hashing MapReduce Job -- 3.2 Document Clustering MapReduce Job -- 4 Experiments and Results -- 4.1 Methodology -- 4.2 Environment and Datasets -- 4.3 Results -- 5 Conclusion -- References -- Impact of Textual Data Augmentation on Linguistic Pattern Extraction to Improve the Idiomaticity of Extractive Summaries -- 1 Context and Objectives -- 2 Background -- 2.1 From Terminology to Patterns and Constructions -- 2.2 Text Summarisation -- 3 Methodology -- 3.1 Pre-training the Original CamemBERT on a Financial Dataset -- 3.2 Fine-Tuning of Our Model on Extractive Summarisation -- 3.3 Heads with Linguistic Knowledge -- 3.4 Data Augmentation in CamemBERT -- 4 Results and Evaluation -- 5 Conclusion -- References -- Explainability in Irony Detection -- 1 Introduction -- 2 Methods -- 3 Experiments and Results -- 4 Conclusion -- References -- Efficient Graph Analytics in Python for Large-Scale Data Science -- 1 Introduction -- 2 Related Work -- 3 Definitions -- 4 A Graph Analytics Function Based on a Semiring -- 4.1 A General Graph Analytics Function. ; 4.2 System Architecture for Function Processing -- 5 Experimental Evaluation -- 6 Conclusions -- References -- Machine Learning and Deep Learning -- A New Accurate Clustering Approach for Detecting Different Densities in High Dimensional Data -- 1 Introduction -- 2 DEnsity-Based Clustering Using WAsserstein Distance -- 2.1 Graph-Oriented Modeling of the Dataset -- 2.2 Probability Density Estimation -- 2.3 Graph Division -- 2.4 Agglomeration of Sub-clusters -- 2.5 Making Hyperparameters Practical -- 3 Experiments -- 3.1 Experiment Protocol -- 3.2 Results and Discussions -- 4 Conclusion -- References -- ODCA: An Outlier Detection Approach to Deal with Correlated Attributes -- 1 Introduction -- 2 Related Works -- 3 Problem Formulation -- 3.1 Learning -- 3.2 Estimating -- 3.3 Combining -- 4 Experiments -- 4.1 Synthetic Dataset -- 4.2 Real Dataasets -- 5 Conclusions -- References -- A Novel Neurofuzzy Approach for Semantic Similarity Measurement -- 1 Introduction -- 2 State-of-the-Art -- 3 A Novel Neurofuzzy Approach for Semantic Similarity -- 4 Experimental Study -- 4.1 Datasets and Evaluation Criteria -- 4.2 Configuration -- 4.3 Results -- 5 Conclusions and Future Work -- References -- Data Warehouse Processes and Maintenance -- Integrated Process Data and Organizational Data Analysis for Business Process Improvement -- 1 Introduction -- 2 Running Example -- 3 Integrating Process Data and Organizational Data -- 4 Data Warehouse Approach -- 5 Example of Application -- 6 Related Work -- 7 Conclusions -- References -- Smart-Views: Decentralized OLAP View Management Using Blockchains -- 1 Introduction -- 2 Smart Views -- 3 Experiments -- 4 Related Work -- 5 Conclusions -- References -- A Workload-Aware Change Data Capture Framework for Data Warehousing -- 1 Introduction -- 2 Problem Analysis -- 3 Proposed Workload-Aware CDC Framework -- 4 Experiments. ; 5 Conclusions and Future Work -- References -- Machine Learning and Analtyics -- Motif Based Feature Vectors: Towards a Homogeneous Data Representation for Cardiovascular Diseases Classification -- 1 Introduction -- 2 Application Domain and Formalism -- 3 Motif-Based Feature Vector Generation -- 4 Evaluation -- 5 Conclusion -- References -- Filter-Based Feature Selection Methods for Industrial Sensor Data: A Review -- 1 Introduction -- 1.1 Background -- 1.2 Previous Surveys -- 2 Supervised Feature Selection Methods -- 3 Experiments and Discussion -- 4 Conclusion -- References -- A Declarative Framework for Mining Top-k High Utility Itemsets -- 1 Introduction -- 2 Preliminaries -- 3 Computing Top-k High Utility Itemsets Using SAT -- 4 Experimental Results -- 5 Conclusion -- References -- Multi-label Feature Selection Algorithm via Maximizing Label Correlation-Aware Relevance and Minimizing Redundance with Mutation Binary Particle Swarm Optimization -- 1 Introduction -- 2 A Novel Multi-label Feature Selection Method -- 2.1 Preliminaries -- 2.2 Feature Selection Criterion Based on Symmetrical Uncertainty -- 2.3 Multi-label Feature Selection Algorithm with Binary Particle Swarm Optimization with Mutation Operation -- 3 Experiments and Analysis -- 3.1 Four Benchmark Data Sets and Two Evaluation Metrics -- 3.2 Experimental Results and Analysis -- 4 Conclusions -- References -- Mining Partially-Ordered Episode Rules with the Head Support -- 1 Introduction -- 2 Problem Definition -- 3 The POERMH Algorithm -- 4 Experiments -- 5 Conclusion -- References -- Boosting Latent Inference of Resident Preference from Electricity Usage - A Demonstration on Online Advertisement Strategies -- 1 Introduction -- 2 The DMAR Model -- 2.1 Problem Definition -- 2.2 Framework -- 3 Experiments -- 4 Conclusions -- References -- Author Index.
• Titolo autorizzato: Big data analytics and knowledge discovery
• ISBN: 3-030-86534-7
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 996464515203316
• Lo trovi qui: Univ. di Salerno
• Serie: Lecture notes in computer science ; ; 12925.