
Machine Learning and Knowledge Discovery in Databases
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
More details
Other editions
Additional editions

Persons
Content
- Intro
- Foreword to the ECML PKDD 2015 Industry,Governmental, and NGO Track
- Foreword to the ECML PKDD 2015 Nectar Track
- Foreword to the ECML PKDD 2015 Demo Track
- Preface
- Organization
- Invited Talks Abstracts(Industrial Track)
- AI Research at Huawei Technologies
- Algorithmic Fashion
- Deep Medical Learning
- Contents - Part III
- Industrial Track
- Autonomous HVAC Control, A Reinforcement Learning Approach
- 1 Introduction
- 2 Background Research
- 3 Markov Decision Processes and Learning Methods
- 3.1 Markov Decision Processes
- 3.2 Reinforcement Learning
- 3.3 Bayesian Inference
- 4 HVAC Control
- 4.1 Occupancy Prediction
- 4.2 HVAC Control Using Q-learning
- 5 Initial Results
- 5.1 Occupancy Prediction
- 5.2 Autonomous Thermostat Control via Q-learning
- Simulation Environment.
- Online Q-learning vs HVAC ``Always On''.
- Offline Q-learning Comfort Analysis.
- Offline Q-learning vs HVAC ``Programmable Control''.
- 6 Conclusions and Future Work
- References
- Clustering by Intent: A Semi-Supervised Method to Discover Relevant Clusters Incrementally
- 1 Introduction
- 2 CBI Methods
- 3 Experiments
- 3.1 Experiment Protocol
- 3.2 Results
- 4 Related Work
- 5 Conclusions and Future Work
- References
- Country-Scale Exploratory Analysis of Call Detail Records Through the Lens of Data Grid Models
- 1 Introduction
- 2 Related Work
- 3 Impacts on Economic Strategy
- 4 Exploratory Analysis through Data Grid Models
- 4.1 Data Grid Exploitation and Visualization
- 5 Exploration Results
- 5.1 Analysis of Call Network between Antennas
- 5.2 Temporal Analysis of the Calls Distribution
- 5.3 Analysis of Output Communications w.r.t. Week Day and Hour
- 5.4 User Mobility Analysis w.r.t. Week Day and Hour
- 6 Conclusion
- References
- Early Detection of Fraud Storms in the Cloud
- 1 Introduction
- 2 Methods
- 2.1 The Problem Setting
- 2.2 The Labels
- 2.3 The Analyzed Signals
- 2.4 Learning Approaches
- 2.5 Performance Evaluation
- 3 Results
- 3.1 Analysis Approaches Comparison
- 3.2 Handling Missing and Corrupted Data
- 3.3 The Effect of the Data Accumulation Period Length
- 4 The System
- 5 Discussion
- References
- Flexible Sliding Windows for Kernel Regression Based Bus Arrival Time Prediction
- 1 Introduction
- 2 Problem Definition
- 2.1 Kernel Regression
- 2.2 Problem Definition
- 3 A Brute-Force Algorithm
- 4 An Approximation Algorithm
- 4.1 Approximation Algorithm
- 4.2 Theoretical Analysis
- 4.3 Further Optimization
- 5 Experiments
- 5.1 Dataset and Experiment Settings
- 5.2 Prediction Accuracy
- 5.3 Effectiveness of Approximation
- 5.4 Interpretation of the Results
- 6 Related Work
- 7 Conclusion and Future Work
- References
- Learning Detector of Malicious Network Traffic from Weak Labels
- 1 Introduction
- 2 Malware Detection
- 3 Multiple Instance Learning of Neyman Pearson Detector
- 3.1 Averaged Stochastic Gradient Descent
- 3.2 Baseline SVM Detector
- 4 Specification of the Datasets
- 5 Experiments
- 5.1 Main Results
- 5.2 Why Does MIL Work Better than SVM?
- 6 Conclusions
- References
- Online Analysis of High-Volume Data Streams in Astroparticle Physics
- 1 Introduction
- 2 Data Analysis in Cherenkov Astronomy
- 2.1 From Raw Data Acquisition to Spectral Analysis
- 2.2 Feature Extraction for Machine Learning
- 2.3 Signal Separation and Energy Estimation: Machine Learning
- 2.4 The Interdisciplinary Gap in Process Development
- 3 Online Data Analysis for the FACT Telescope
- 3.1 The streams Data Flow Framework
- 3.2 FACT-Tools: Processing Telescope Data
- 3.3 Integration of Machine Learning Libraries
- 4 Experiments
- 4.1 Gamma-Hadron Separation with WEKA Classifiers
- 4.2 Signal Separation with Local Models
- 4.3 Throughput Performance of the FACT-Tools
- 5 Summary and Conclusion
- References
- Robust Representation for Domain Adaptation in Network Security
- 1 Introduction
- 2 Problem Statement
- 3 Invariant Representation of Bags
- 3.1 Shift Invariance with Self-Similarity Matrix
- 3.2 Scale Invariance with Local Feature Normalization
- 3.3 Permutation and Size Invariance with Histograms
- 4 Online Similarity Learning
- 5 Application in Network Security
- 5.1 Specification of the Datasets
- 5.2 Experimental Evaluation
- 6 Conclusion
- References
- Safe Exploration for Active Learning with Gaussian Processes
- 1 Introduction
- 2 Related Work
- 3 Safe Exploration for Active Learning
- 3.1 Gaussian Processes
- 3.2 Exploration Strategy
- 3.3 Safety Constraint
- 3.4 Safe Active Learning: The Algorithm
- 4 Theoretical Analysis
- 5 Evaluations
- 6 Conclusion and Outlook
- References
- Semi-Supervised Consensus Clustering for ECG Pathology Classification
- 1 Introduction
- 2 ECG Pathology Classification
- 3 Semi-Supervised Consensus Clustering (SSCC)
- 4 Experiments
- 4.1 Data Description
- 4.2 Feature Extraction
- 4.3 Setup
- 4.4 Validation of the Algorithm: Benchmark Datasets
- 4.5 Application on Real Datasets: ECG MIT-BIH Database
- 5 Conclusions
- References
- Two Step Graph-Based Semi-supervised Learning for Online Auction Fraud Detection
- 1 Introduction
- 2 Data Description
- 2.1 Resources
- 2.2 Graph Construction
- 3 Proposed Approach
- 3.1 Problem Definition
- 3.2 Modified Adsorption (MAD)
- 3.3 MAD for Online Auction Fraud Detection
- 3.4 2-STEP Model
- 4 Experiments
- 4.1 Evaluation Metric
- 4.2 Methodology
- 4.3 Results
- 5 Discussion
- 6 Related Work
- 7 Conclusion
- References
- Watch-It-Next: A Contextual TV Recommendation System
- 1 Introduction
- 2 Background
- 2.1 Building Blocks
- 2.2 Related Work
- 3 Algorithms
- 3.1 Memory-Based Popularity Baselines
- 3.2 Collaborative Filtering Methods
- 3.3 Contextual Personalization
- 4 Experimental Settings
- 4.1 The Data
- 4.2 Emulating Inventories
- 4.3 Evaluation Metric
- 5 Experimental Results
- 5.1 Attributing Context to Users
- 5.2 The Exploratory Setting
- 5.3 The Habitual Setting
- 6 Conclusions and Future Work
- References
- Nectar Track
- Bayesian Hypothesis Testing in Machine Learning
- 1 Introduction
- 2 Software
- References
- Data-Driven Exploration of Real-Time Geospatial Text Streams
- 1 Introduction and Problem Definition
- 2 Stream-Enabled X-means Clustering for Textual Data
- 3 Orchestrating Filters Based on Classification
- 4 Conclusion
- References
- Discovering Neutrinos Through Data Analytics
- 1 Introduction
- 2 Selection of Neutrinos Events
- 2.1 Data Preprocessing
- 2.2 Variable Selection
- 2.3 Performance of the Random Forest
- 3 Unfolding and the Resulting Energy Spectrum
- 4 Summary and Results
- References
- Listener-Aware Music Recommendation from Sensor and Social Media Data
- 1 Introduction
- 2 Automatic Playlist Adaptation Based on Sensor Data
- 3 Music Recommendation Based on Social Media Mining
- 4 Outlook
- References
- Logic-Based Incremental Process Mining
- 1 Introduction
- 2 A FOL-Based Proposal
- 3 Evaluation
- 4 Conclusions
- References
- Mobility Mining for Journey Planning in Rome
- 1 Introduction
- 2 PETRA System Components
- 2.1 Data Management
- 2.2 Mobility Mining
- 2.3 The Multi-modal Journey Planner
- 3 Case Study
- 3.1 Rome Data
- 3.2 Importing Rome's Data
- 3.3 Impact of Routines in Journey Planning
- 4 Conclusions
- References
- Pattern Structures and Concept Lattices for Data Mining and Knowledge Processing
- 1 Pattern Structures in Formal Concept Analysis
- 2 Applications
- References
- Privacy Preserving Blocking and Meta-Blocking
- 1 Introduction
- 2 Privacy Preserving Blocking and Meta-Blocking
- 3 Conclusion and Future Work
- References
- Social Data Mining and Seasonal Influenza Forecasts: The FluOutlook Platform
- 1 Introduction
- 2 Methodology
- 2.1 Simulation Engine
- 2.2 Tracking Seasonal Flu with Social Data Mining
- Inferring Initial Infections from Twitter.
- Inferring Initial Infections from Influweb.
- 2.3 Generative Model Selection and Forecast Output
- 3 Conclusion
- References
- Star Classification Under Data Variability: An Emerging Challenge in Astroinformatics
- 1 Introduction
- 2 Cepheid Variable Star Classification
- 3 Conclusions and Remarks
- References
- The Evolution of Social Relationships and Strategies Across the Lifespan
- References
- Understanding Where Your Classifier Does (Not) Work
- 1 Introduction
- 2 Related Work
- 3 The SCaPE Model Class for EMM
- 4 Experimental Results
- 5 Conclusions
- References
- Visual Analytics Methodology for Scalable and Privacy-Respectful Discovery of Place Semantics from Episodic Mobility Data
- 1 Introduction
- 2 Problem Statement and Methodology Overview
- 3 Feasibility Studies
- References
- Will This Paper Increase Your h-index?
- References
- Demo Track
- CardioWheel: ECG Biometrics on the Steering Wheel
- 1 Introduction
- 2 Target Users
- 3 System Overview
- 4 Demonstration Prototype
- References
- Data Patterns Explained with Linked Data
- 1 Introduction
- 2 Dedalo's Implementation
- 3 Demo Scenarios
- References
- Gazouille: Detecting and Illustrating Local Events from Geolocalized Social Media Streams
- 1 Introduction
- 2 System Overview
- 3 Use Case: New York City
- References
- HiDER: Query-Driven Entity Resolution for Historical Data
- 1 Introduction
- 2 The HiDER System
- 3 Concluding Remarks
- References
- IICE: Web Tool for Automatic Identification of Chemical Entities and Interactions
- 1 Introduction
- 2 Architecture Overview
- 3 Web Tool
- References
- Interactively Exploring Supply and Demand in the UK Independent Music Scene
- 1 Introduction and Motivation
- 2 Data Collection
- 3 Overview of Interface
- 4 Conclusions
- References
- Monitoring Production Equipment in Remote Locations
- 1 Objectives
- 2 Project Implementation
- 3 Results
- 4 Innovative Aspe ects of the Project
- 5 Future Work
- MultiNets: Web-Based Multilayer Network Visualization
- 1 Introduction
- 2 Implementation
- 3 Summary
- References
- OMEGA: An Order-Preserving SubMatrix Mining, Indexing and Search Tool
- 1 Introduction
- 2 The Architecture and Key Technologies of OMEGA
- 3 Demonstration
- References
- Probabilistic Programming in Anglican
- 1 Introduction
- 2 Design Outline
- 3 Usage Patterns
- 4 Anglican Examples
- References
- ProbLog2: Probabilistic Logic Programming
- 1 Introduction
- 2 Language
- 3 System Blocks
- 4 System Usage
- References
- Real Time Detection and Tracking of Spatial Event Clusters
- Abstract.
- 1 Problem Setting
- 2 Approach
- 3 Example
- References
- S&P360: Multidimensional Perspective on Companies from Online Data Sources
- 1 Introduction
- 2 Architecture
- 2.1 Data Layer
- 2.2 Analytics
- 2.3 API
- 2.4 User Interface Layer
- 3 Demo on Real World Data
- 4 Video and Requirements
- References
- Scavenger -- A Framework for Efficient Evaluation of Dynamic and Modular Algorithms
- 1 Introduction
- 2 The Scavenger Framework
- 3 Case Study: Autoencoders
- 4 Conclusion
- References
- UrbanHubble: Location Prediction and Geo-Social Analytics in LBSN
- 1 Introduction
- 2 UrbanHubble Tool
- 2.1 Architecture
- 2.2 Related Works
- 3 Conclusions
- References
- VIPER -- Visual Pattern Explorer
- 1 Introduction
- 2 Related Work
- 3 VIPER -- Visual Pattern Explorer
- 3.1 Preliminary User Study: Movie Ratings
- 4 Conclusions
- References
- Visualization Support to Interactive Cluster Analysis
- 1 Introduction
- 2 Interactive Two-Way Cluster Analysis of Spatial Time Series
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.