Introduction | Content | Author | Resource

Introduction

Temporal Intelligence—An Artificial Intelligence Approach for Time Series Analysis

This book is intended for readers interested in Time Series Analysis and Artificial Intelligence. It aims to systematically review relevant basic knowledge and provide in-depth explanations of key techniques and practical applications, ranging from traditional statistical modeling to cutting-edge deep learning methods. The author team will continuously monitor open-source communities and academic frontiers, carefully considering feedback from experts and scholars. We plan to provide regular updates, striving to create a monograph on time series intelligence that is readable, practical, and of high academic value.

The book consists of nine parts: Introduction, Basic Concepts, Time Series Forecasting, Time Series Anomaly Detection, Time Series Classification, Automated Time Series Analysis, Time Series Foundation Models, Time Series Benchmarks, and Neural Differential Equations for Time Series Analysis. Each chapter is based on the author team's exploration and practice in related fields and represents solely our own understanding. We sincerely welcome valuable feedback from readers regarding any omissions or errors. In the future, the team will continue to expand upon cutting-edge methods, practical tools, and evaluation standards to gradually refine the content of this monograph.

The current complete PDF version of this book is available at Time Series Intelligence.pdf. Alternative download links: Google Drive, Baidu Netdisk. The table of contents for each chapter is shown below.

Content

Chapter 1: Introduction

1.1 Target Audience

1.2 Structure of the Book

1.3 Getting Started with Time Series

1.4 Definition and Classification of Time Series

1.5 Evolution of Time Series Algorithms

Chapter 2: Basic Concepts

2.1 Overview of Time Series Data

2.2 Time Series Data Governance

2.3 Perspectives on Time Series Modeling

2.4 Basic Deep Learning Models

Chapter 3: Time Series Forecasting

3.1 Definition and Process of Forecasting

3.2 Training and Evaluation of Forecasting Models

3.3 Time Series Forecasting Models

3.4 Channel Relationships in Time Series

3.5 Probabilistic Time Series Forecasting

Chapter 4: Time Series Anomaly Detection

4.1 Definition and Process of Anomaly Detection

4.2 Types of Time Series Anomalies

4.3 Training and Evaluation of Anomaly Detection Models

4.4 Time Series Anomaly Detection Models

Chapter 5: Time Series Classification

5.1 Definition and Process of Classification

5.2 Training and Evaluation of Classification Models

5.3 Time Series Classification Models

5.4 Self-Supervised Methods for Classification

Chapter 6: Automated Time Series Analysis

6.1 Introduction to Automated Analysis

6.2 Automated Model Selection

6.3 Automated Model Ensembling

6.4 Automated Model Design

Chapter 7: Time Series Foundation Models

7.1 Pre-training-based Foundation Models

7.2 LLM-based Foundation Models

Chapter 8: Time Series Benchmarks

8.1 Background and Significance of Benchmarks

8.2 Development History of Benchmarks

8.3 Framework of Time Series Benchmarks

8.4 Data Layer

8.5 Method Layer

8.6 Evaluation Layer

8.7 Reporting Layer

Chapter 9: Neural Differential Equations for Time Series

9.1 Differential Equations and Time Series

9.2 Neural Differential Equations (Neural ODEs)

9.3 Related Application Research

Author

Editor-in-Chief

Chenjuan Guo
School of Data Science and Engineering, East China Normal University
Professor, PhD Supervisor, National-level Young Talent, Shanghai Leading Talent

Associate Editors

Bin Yang

School of Data Science and Engineering, East China Normal University

Professor (Level 2), PhD Supervisor, National-level Leading Talent

Jilin Hu

School of Data Science and Engineering, East China Normal University

Professor, PhD Supervisor, National-level Young Talent

Yang Shu

School of Data Science and Engineering, East China Normal University

Assistant Professor, Master's Supervisor, Chenhui Scholar

Core Writing Team

The compilation of this book was jointly completed by members of the Decision Intelligence Lab, East China Normal University.
They were responsible for chapter writing, case design, experimental verification, and the development of supplementary resources.
Participating members are as follows (in alphabetical order):
Peng Chen, Yuxuan Chen, Hanyin Cheng, Yifei Ding, Hongfan Gao, Shiyan Hu, Shanshan Huang, Jianxin Jin, Zhi Lei, Beibu Li, Zhe Li, Zhengyu Li, Junkai Lu, Yuning Lu, Xiangfei Qiu, Yuying Qiu, Qichao Shentu, Jindong Tian, Yihang Wang, Linfeng Wang, Siyuan Wang, Xingjian Wu, Siyu Yan, Ronghui Xu, Ruitong Zhang, Xingze Zheng

Resource

[GitHub] [Lab] [Mail]