New frontiers in mining complex patterns : 8th International Workshop, NFMCP 2019, held in Conjunction with ECML-PKDD 2019, Würzburg, Germany, September 16, 2019, Revised selected papers / Michelangelo Ceci, Corrado Loglisci, Giuseppe Manco, Elio Masciari, Zbigniew Ras (eds.).

Print version record.

Intro -- Preface -- Organization -- Effective Linear Models for Learning with Sequences and Time Series (Abstract of Invited Talk) -- Contents -- Complex Patterns -- A Framework for Pattern Mining and Anomaly Detection in Multi-dimensional Time Series and Event Logs -- 1 Introduction -- 2 Preliminaries -- 3 Method -- 3.1 Time Series Representation for Pattern Mining -- 3.2 Pattern Mining -- 3.3 Pattern-Based Anomaly Detection -- 3.4 Implementation of the Framework -- 4 Use Cases -- 5 Related Work -- 6 Conclusion -- References

A Heuristic Approach for Sensitive Pattern Hiding with Improved Data Quality -- 1 Introduction -- 2 Related Work: Heuristic Based Algorithms -- 3 Background -- 3.1 Basic Terminologies -- 3.2 Problem Statement -- 4 Proposed Solution: MinMax Algorithm -- 4.1 Time Complexity Analysis -- 4.2 Example -- 5 Experimental Results -- 5.1 Varying Percentage of Sensitive Itemsets -- 5.2 Varying Minimum Support Threshold -- 6 Conclusion -- References -- Classification and Regression -- Interpretable Survival Gradient Boosting Models with Bagged Trees Base Learners -- Abstract -- 1 Introduction

2 Survival Analysis -- 2.1 Notations -- 2.2 Partial Likelihood -- 2.3 Cox Proportional Hazards Model -- 3 Proposed Method -- 3.1 Gradient Boosting -- 3.2 Additive Representations -- 3.3 Base Learner Function -- 3.4 Loss Function -- 4 Evaluation -- 4.1 Datasets -- 4.2 Methods -- 4.3 Parameters -- 4.4 Concordance Index -- 4.5 Validation -- 5 Results and Discussion -- 5.1 Base Learners -- 5.2 Interpretability -- 5.3 Feature Selection -- 6 Conclusions -- References -- Neural Hybrid Recommender: Recommendation Needs Collaboration -- 1 Introduction -- 2 Neural Hybrid Recommender -- 3 Experiments

3.1 Datasets -- 3.2 Handling Text Data -- 3.3 Evaluation Process -- 3.4 Negative Sampling -- 3.5 Baselines -- 3.6 Parameter Setting -- 3.7 Performance Results -- 4 Conclusion -- References -- Discovering Discriminative Nodes for Classification with Deep Graph Convolutional Methods -- 1 Context and Motivation -- 2 Related Work -- 3 Interpreting Graph Convolutional Network Models with Grad-CAM -- 3.1 Graph Sparsification -- 3.2 Deep Graph Convolutional Neural Networks -- 3.3 DGCNN Interpretability -- 4 Experimental Evaluation and Results -- 4.1 Data Generation

4.2 Classification Performance Evaluation -- 4.3 Interpretability Heatmaps -- 5 Discussion -- 6 Conclusion -- References -- Streams and Times Series -- Soft Voting Windowing Ensembles for Learning from Partially Labelled Streams -- 1 Introduction -- 2 Background -- 3 LESS-TWE Online Learning -- 3.1 Hybrid Sliding-Tumbling Windows -- 3.2 Weighted Soft Voting -- 3.3 Online Labelling -- 3.4 Unlabelled Drift Detection -- 4 Experimental Evaluation -- 4.1 Benchmark Data Sets -- 4.2 Effects of Training with a Lower Percentage of Labelled Data -- 4.3 Comparison in Terms of Accuracy and Runtime

4.4 Intrusion Detection Databases

Includes author index.

This book constitutes the refereed post-conference proceedings of the 8th International Workshop on New Frontiers in Mining Complex Patterns, NFMCP 2019, held in conjunction with ECML-PKDD 2019 in Würzburg, Germany, in September 2019. The workshop focused on the latest developments in the analysis of complex and massive data sources, such as blogs, event or log data, medical data, spatio-temporal data, social networks, mobility data, sensor data and streams.