Intro
Preface
Conference Organization
Contents
Data Mining Methods and Applications
Mining Partially-Ordered Episode Rules in an Event Sequence
1 Introduction
2 Problem Definition
3 The POERM Algorithm
4 Experimental Evaluation
5 Conclusion
References
Investigating Crossover Operators in Genetic Algorithms for High-Utility Itemset Mining
1 Introduction
2 Preliminaries
3 The Headless Chicken Test for HUIM Using GAs
4 Experiments and Results
4.1 Runtime
4.2 Discovered HUIs
4.3 Convergence
5 Conclusion
References

Complexes of Low Dimensional Linear Classifiers with L1 Margins
1 Introduction
2 Linear Separability of Feature Vectors
3 Dual Hyperplanes and Vertices in the Parameter Space
4 Perceptron Criterion Function
5 Basis Exchange Algorithms
6 Reduced Criterion Functions
7 Complexes of Linear Classifiers
8 Concluding Remarks
References
Automatic Identification of Bird Species from Audio
1 Introduction
2 State of Art
2.1 Preprocessing and Feature Extraction
2.2 Deep Learning for Sound Classification
2.3 Other Approaches
3 Our Approach
3.1 Pre-processing

3.2 Feature Extraction
3.3 Deep Learning Model Architectures
4 Experiments and Evaluation
4.1 Evaluation Metrics
4.2 Experiments
4.3 Best Results
5 Conclusions and Future Work
References
Session Based Recommendations Using Recurrent Neural Networks
Long Short-Term Memory
1 Introduction
2 Methods of Content Recommendation
2.1 Multi-Layer Perceptron
2.2 Autoencoders
2.3 Boltzmann Machines
3 Evaluation of Recommendation Systems
3.1 MAE
3.2 MSE
3.3 Precision
3.4 Recall
4 Using the LSTM as a Recommender System
4.1 Recurrent Neural Networks

4.2 Long Short-Term Memory
4.3 Data Preparation
4.4 Model Definition
5 Results and Discussions
5.1 Experiment
5.2 Recall and Precision Results
6 Conclusions and Future Work
References
UVDS: A New Dataset for Traffic Forecasting with Spatial-Temporal Correlation
1 Introduction
2 Literature Review
2.1 Traffic Forecasting Using Graph Neural Network
2.2 Public Traffic Flow Datasets
3 UVDS: Urban Vehicle Detection System Dataset
3.1 Data Description and Research Challenges
3.2 Graph Construction
4 Baseline Results and Discussion
4.1 Baseline Models

4.2 Experimental Results
4.3 Open Research Issues Using UVDS Dataset
5 Conclusion
References
A Parallelized Frequent Temporal Pattern Mining Algorithm on a Time Series Database
1 Introduction
2 A Frequent Temporal Inter-object Pattern Mining Task
3 The Parallelized Frequent Temporal Inter-object Pattern Mining Algorithm on a Time Series Database
3.1 BranchTree
3.2 From BranchTree to PTP
4 Empirical Evaluation
5 Conclusion
References
An Efficient Approach for Mining High-Utility Itemsets from Multiple Abstraction Levels
1 Introduction
2 Related Work