Online sequential pattern mining and association discovery by advanced artificial intelligence and machine learning techniques

Shian Chang Huang, Chei Chang Chiou, Jui Te Chiang, Cheng Feng Wu

Research output: Contribution to journalArticle

Abstract

With the advances in information science, vast amounts of financial time series data can been collected and analyzed. In modern time series analysis, sequential pattern mining (SPM) and association discovery (AD) are the most important techniques to predict the future trends. This study aims at developing advanced SPM and AD for financial data by cutting edge techniques from artificial intelligence and machine learning. The nonlinearity and non-stationarity of financial time series dynamics pose a major challenge for SPM and AD. This study employs time–frequency analysis to extract features for SPM. Then, a sparse multi-manifold clustering (SMMC) is used to partition the feature space into several disjointed regions for better AD. Finally, local relevance vector machines (RVMs) are employed for AD and perform the forecasting. Different from traditional methods, the novel forecasting system operates on multiple resolutions and multiple dynamic regimes. SMMC finds both the neighbors and the weights automatically by a sparse solution, which approximately spans a low-dimensional affine subspace at that point. RVM, the Bayesian kernel machines, can produce parsimonious models with excellent generalization properties. Taking multiple time series data from financial markets as an example, the empirical results demonstrate that the proposed model outperforms traditional models and significantly reduces the forecasting errors. The framework is effective and suitable for other time series forecasting.

Original languageEnglish
JournalSoft Computing
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Sequential Patterns
Artificial intelligence
Learning systems
Time series
Mining
Artificial Intelligence
Machine Learning
Relevance Vector Machine
Forecasting
Financial Data
Financial Time Series
Time Series Data
Information science
Time series analysis
Clustering
Kernel Machines
Multiple Time Series
Time Series Forecasting
Time-frequency Analysis
Nonstationarity

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Geometry and Topology

Cite this

@article{b39adc9ae74c481aaaeff31d091affa7,
title = "Online sequential pattern mining and association discovery by advanced artificial intelligence and machine learning techniques",
abstract = "With the advances in information science, vast amounts of financial time series data can been collected and analyzed. In modern time series analysis, sequential pattern mining (SPM) and association discovery (AD) are the most important techniques to predict the future trends. This study aims at developing advanced SPM and AD for financial data by cutting edge techniques from artificial intelligence and machine learning. The nonlinearity and non-stationarity of financial time series dynamics pose a major challenge for SPM and AD. This study employs time–frequency analysis to extract features for SPM. Then, a sparse multi-manifold clustering (SMMC) is used to partition the feature space into several disjointed regions for better AD. Finally, local relevance vector machines (RVMs) are employed for AD and perform the forecasting. Different from traditional methods, the novel forecasting system operates on multiple resolutions and multiple dynamic regimes. SMMC finds both the neighbors and the weights automatically by a sparse solution, which approximately spans a low-dimensional affine subspace at that point. RVM, the Bayesian kernel machines, can produce parsimonious models with excellent generalization properties. Taking multiple time series data from financial markets as an example, the empirical results demonstrate that the proposed model outperforms traditional models and significantly reduces the forecasting errors. The framework is effective and suitable for other time series forecasting.",
author = "Huang, {Shian Chang} and Chiou, {Chei Chang} and Chiang, {Jui Te} and Wu, {Cheng Feng}",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s00500-019-04100-5",
language = "English",
journal = "Soft Computing",
issn = "1432-7643",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Online sequential pattern mining and association discovery by advanced artificial intelligence and machine learning techniques

AU - Huang, Shian Chang

AU - Chiou, Chei Chang

AU - Chiang, Jui Te

AU - Wu, Cheng Feng

PY - 2019/1/1

Y1 - 2019/1/1

N2 - With the advances in information science, vast amounts of financial time series data can been collected and analyzed. In modern time series analysis, sequential pattern mining (SPM) and association discovery (AD) are the most important techniques to predict the future trends. This study aims at developing advanced SPM and AD for financial data by cutting edge techniques from artificial intelligence and machine learning. The nonlinearity and non-stationarity of financial time series dynamics pose a major challenge for SPM and AD. This study employs time–frequency analysis to extract features for SPM. Then, a sparse multi-manifold clustering (SMMC) is used to partition the feature space into several disjointed regions for better AD. Finally, local relevance vector machines (RVMs) are employed for AD and perform the forecasting. Different from traditional methods, the novel forecasting system operates on multiple resolutions and multiple dynamic regimes. SMMC finds both the neighbors and the weights automatically by a sparse solution, which approximately spans a low-dimensional affine subspace at that point. RVM, the Bayesian kernel machines, can produce parsimonious models with excellent generalization properties. Taking multiple time series data from financial markets as an example, the empirical results demonstrate that the proposed model outperforms traditional models and significantly reduces the forecasting errors. The framework is effective and suitable for other time series forecasting.

AB - With the advances in information science, vast amounts of financial time series data can been collected and analyzed. In modern time series analysis, sequential pattern mining (SPM) and association discovery (AD) are the most important techniques to predict the future trends. This study aims at developing advanced SPM and AD for financial data by cutting edge techniques from artificial intelligence and machine learning. The nonlinearity and non-stationarity of financial time series dynamics pose a major challenge for SPM and AD. This study employs time–frequency analysis to extract features for SPM. Then, a sparse multi-manifold clustering (SMMC) is used to partition the feature space into several disjointed regions for better AD. Finally, local relevance vector machines (RVMs) are employed for AD and perform the forecasting. Different from traditional methods, the novel forecasting system operates on multiple resolutions and multiple dynamic regimes. SMMC finds both the neighbors and the weights automatically by a sparse solution, which approximately spans a low-dimensional affine subspace at that point. RVM, the Bayesian kernel machines, can produce parsimonious models with excellent generalization properties. Taking multiple time series data from financial markets as an example, the empirical results demonstrate that the proposed model outperforms traditional models and significantly reduces the forecasting errors. The framework is effective and suitable for other time series forecasting.

UR - http://www.scopus.com/inward/record.url?scp=85066863977&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066863977&partnerID=8YFLogxK

U2 - 10.1007/s00500-019-04100-5

DO - 10.1007/s00500-019-04100-5

M3 - Article

AN - SCOPUS:85066863977

JO - Soft Computing

JF - Soft Computing

SN - 1432-7643

ER -