Trends and Applications in Knowledge Discovery and Data Mining : PAKDD 2024 Workshops, RAFDA and IWTA, Taipei, Taiwan, May 7–10, 2024, Proceedings / edited by Zhaoxia Wang, Chang Wei Tan.

Format:

Book

Author/Creator:

Wang, Zhaoxia.

Contributor:

Tan, Chang Wei.

Series:

Lecture Notes in Artificial Intelligence, 2945-9141 ; 14658

Language:

English

Subjects (All):

Artificial intelligence.

Computers.

Data mining.

Information storage and retrieval systems.

Image processing--Digital techniques.

Image processing.

Computer vision.

Artificial Intelligence.

Computing Milieux.

Data Mining and Knowledge Discovery.

Information Storage and Retrieval.

Computer Imaging, Vision, Pattern Recognition and Graphics.

Local Subjects:

Artificial Intelligence.

Computing Milieux.

Data Mining and Knowledge Discovery.

Information Storage and Retrieval.

Computer Imaging, Vision, Pattern Recognition and Graphics.

Physical Description:

1 online resource (185 pages)

Edition:

1st ed. 2024.

Place of Publication:

Singapore : Springer Nature Singapore : Imprint: Springer, 2024.

Summary:

This book constitutes the workshops that have been held in conjunction with the 27th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2023, which took place in Osaka, Japan, during May 25–28, 2023. For RAFDA 2024, Workshop on Research and Applications of Foundation Models for Data Mining and Affective Computing, 15 submissions have been received and 9 full papers have been accepted for publication. For IWTA 2024, International Workshop on Temporal Analytics, 4 full papers have been accepted from a total of 6 submissions. .

Contents:

Intro

RAFDA 2024 Preface

RAFDA 2024 Organization

IWTA 2024 Preface

IWTA 2024 Organization

Contents

PAKDD 2024 workshop on Research and Applications of Foundation Models for Data Mining and Affective Computing (RAFDA 2024)

Evaluation of Orca 2 Against Other LLMs for Retrieval Augmented Generation

1 Background and Introduction

2 Related Work

3 Methodology

3.1 Workflow Overview

3.2 Gathering and Processing Data

3.3 Assessment and Comparative Evaluation of Orca 2 Against Other LLMs

3.4 Assessment Criteria

3.5 Experiment Setup

4 Experiments Results

4.1 LLM Generation Quality

4.2 LLM Inference Speed

5 Conclusions

A Appendix

References

Toward Interpretable Graph Classification via Concept-Focused Structural Correspondence

1 Introduction

2.1 Interpretable Graph Neural Networks

2.2 Graph Structure Similarity Measurement

2.3 Deep Learning for Case-Based Reasoning

3.1 Problem Formulation

3.2 Graph Structure Correspondence

3.3 Representation Learning and Node Importance Weighting

3.4 Interpretable Non-parametric Predictor

3.5 Explanation Methods

3.6 Computational Complexity

4 Experiments

4.1 Baselines and Datasets

4.2 Implementations and Configurations

4.3 Performance Comparison with Baselines

4.4 Ablation Studies on Predictive Performance

4.5 Interpretation Analysis

4.6 User Perception of Explanations

5 Conclusion and Future Work

InteraRec: Interactive Recommendations Using Multimodal Large Language Models

3 The InteraRec Framework

3.1 Screenshot Generation

3.2 Behavioral Summarization

3.3 Response Generation

4 Illustration

4.1 Assortment Planning

4.2 Multinomial Logit (MNL)

4.3 LLM

4.4 Illustrative Examples

5 Conclusion.

A Appendix

Research on Dynamic Community Detection Method Based on Multi-dimensional Feature Information of Community Network

3 Problem Description

4 Algorithmic Model

4.1 Feature Processing

4.2 Reference to Historical Information

5 Experiments

5.1 Datasets

5.2 Evaluation Indicators

5.3 Comparison Algorithm

5.4 First Experiment

5.5 Second Experiment

6 Conclusion

From Tweets to Token Sales: Assessing ICO Success Through Social Media Sentiments

2 Related Work and Motivations

3 Hypotheses

4 Dataset

5 Experiment and Methodology

5.1 Data Processing

5.2 Machine Learning Methods

6 Results and Discussion

6.1 Individual Hypothesis Testing

6.2 Combined Hypotheses Testing

6.3 Further Discussion

7 Conclusion

Enhanced Graph Neural Network for Session-Based Recommendation with Static and Dynamic Information

3 Preliminary

3.1 Problem Statement

3.2 Graph Construction

4 Methodology

4.1 Static and Dynamic Information Enhanced Item Representation Learning Layer (SDI)

4.2 Session Representation Learning Layer

4.3 Prediction Layer

5.1 Experimental Settings

5.2 Overall Comparison (RQ1)

5.3 Ablation Study (RQ2)

5.4 Impact of Aggregation Operations (RQ3)

5.5 Impact of SIL Depth (RQ4)

Construction of Academic Innovation Chain Based on Multi-level Clustering of Field Literature

2 Methodology

2.1 Overview of Methodology

2.2 Multi-level Text Feature Mining Algorithms

2.3 Kmeans Clustering Algorithm

2.4 Innovation Points Extraction

3 Empirical Research

3.1 Data Collection and Preprocessing

3.2 Multi-level Clustering Experiments.

3.3 Innovation Points Extraction Experiments

3.4 Constraction of Academic Innovation Chain

3.5 Comparison Experiment

4 Discussion

DLVS4Audio2Sheet: Deep Learning-Based Vocal Separation for Audio into Music Sheet Conversion

3.1 Overall Design of the DLVS4Audio2Sheet Method

3.2 Vocal Separation Leveraging the Two Deep Learning Models: Open-Unmix and BSRNN

3.3 Audio to Score Sheet Through MIDI File

4.1 Scarcity of Datasets

4.2 Raw Datasets

4.3 Data Augmentation

5 Experimentation and Results

5.1 Vocal Separation Results and Comparisons

5.2 Converting Audio to Score Sheet Through MIDI File

6 Conclusion, Limitations and Future Works

6.1 Conclusion

6.2 Limitations and Future Works

Explainable AI for Stress and Depression Detection in the Cyberspace and Beyond

2 Data Collection

3 Data Analysis

3.1 Concept Parsing

3.2 Subjectivity Detection

3.3 Polarity Classification

3.4 Intensity Ranking

3.5 Emotion Recognition

3.6 Aspect Extraction

3.7 Personality Prediction

3.8 Sarcasm Identification

3.9 Depression Categorization

3.10 Toxicity Spotting

3.11 Engagement Measurement

3.12 Well-Being Assessment

4 Results

5 Conclusion

International Workshop on Temporal Analytics (IWTA 2024)

Finding Foundation Models for Time Series Classification with a PreText Task

2.1 Deep Learning Techniques

2.2 Pre-training Deep Learning Techniques

3 Proposed Method

3.1 Pretext Task

4 Results and Analysis

4.1 Comparing Pre-Training with Baseline (Ensemble)

4.2 Visualizing the Filters

4.3 Comparison with the State-of-the-Art

References.

Next Item and Interval Prediction of New Users Using Meta-Learning on Dynamic Network

2.1 Graph-Based Sequence Recommendation Model

2.2 Meta-Learning for Cold-Start Problem

3.1 Problem Definition

3.2 Model Architecture

4 Evaluations

4.1 Experimental Setup

4.2 Experimental Result

Adaptive Knowledge Sharing in Multi-Task Learning: Insights from Electricity Data Analysis

2 Problem Definition

3 Framework

3.1 Tasks Learning Design

3.2 Two-Phase Training

4.2 Experimental Results

5 Related Works

6 Conclusions

Handling Concept Drift in Non-stationary Bandit Through Predicting Future Rewards

2 Background and Related Works

3.2 Time-Series Concept Drift Adaptive Algorithm

4.1 Dataset

4.2 Baseline

4.3 Experiment Results

Author Index.

Notes:

Description based on publisher supplied metadata and other sources.

ISBN:

9789819726509

9819726506

OCLC:

1432252858