Grammar-Based Feature Generation for Time-Series Prediction

1. Front Matter

   Pages i-xi

   PDF

2. Introduction

   • Anthony Mihirana De Silva, Philip H. W. Leong

   Pages 1-11

3. Feature Selection

   • Anthony Mihirana De Silva, Philip H. W. Leong

   Pages 13-24

4. Grammatical Evolution

   • Anthony Mihirana De Silva, Philip H. W. Leong

   Pages 25-33

5. Grammar Based Feature Generation

   • Anthony Mihirana De Silva, Philip H. W. Leong

   Pages 35-50

6. Application of Grammar Framework to Time-Series Prediction

   • Anthony Mihirana De Silva, Philip H. W. Leong

   Pages 51-62

7. Case Studies

   • Anthony Mihirana De Silva, Philip H. W. Leong

   Pages 63-83

8. Conclusion

   • Anthony Mihirana De Silva, Philip H. W. Leong

   Pages 85-87

9. Back Matter

   Pages 89-99

   PDF

This book proposes a novel approach for time-series prediction using machine learning techniques with automatic feature generation. Application of machine learning techniques to predict time-series continues to attract considerable attention due to the difficulty of the prediction problems compounded by the non-linear and non-stationary nature of the real world time-series. The performance of machine learning techniques, among other things, depends on suitable engineering of features. This book proposes a systematic way for generating suitable features using context-free grammar. A number of feature selection criteria are investigated and a hybrid feature generation and selection algorithm using grammatical evolution is proposed. The book contains graphical illustrations to explain the feature generation process. The proposed approaches are demonstrated by predicting the closing price of major stock market indices, peak electricity load and net hourly foreign exchange client trade volume. The proposed method can be applied to a wide range of machine learning architectures and applications to represent complex feature dependencies explicitly when machine learning cannot achieve this by itself. Industrial applications can use the proposed technique to improve their predictions.

• Context-Free Grammar
• Feature Generation
• Feature Selection
• Grammatical Evolution
• Time-Series Prediction
• quantitative finance

• Electrical and Information Engineering, University of Sydney, Sydney, Australia

  Anthony Mihirana De Silva

• Electrical and Information Engineering, University of Sydney, East Killara, Australia

  Philip H. W. Leong

Policies and ethics