Enhanced symbolic aggregate approximation method for financial time series data representation

Peiman Mamani Barnaghi, Azuraliza Abu Bakar, Zulaiha Ali Othman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Data representation is one of the most important tasks in time series data pre-processing. Time series data representation is required to make the data more suitable for data mining specifically for prediction. Time series data is characterized by its numerical and continuous values. One of the data representation methods for time series is the Symbolic Aggregate Approximation (SAX) which uses mean values as the basis of representation of the data. However. representing the time series financial data with the mean value often causes the loss of patterns that can describes important pieces of information. The aim of this study is to propose an enhancement of SAX representation purposely for the financial time series data. The Enhanced SAX (EN-SAX) adds two new values to the original mean value for each segment in SAX. These values enable better representation for each segment in a lower dimension and keep some of the important patterns that are meaningful in financial time series data. The experimental results show that the EN-SAX representation manages to give lower error rates compared to SAX and improves the prediction accuracy.

Original languageEnglish
Title of host publicationProceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012
Pages790-795
Number of pages6
Publication statusPublished - 2012
Event2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012 - Taipei
Duration: 23 Oct 201225 Oct 2012

Other

Other2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012
CityTaipei
Period23/10/1225/10/12

Fingerprint

Time series
Data mining
Processing

Keywords

  • dimensionality reduction
  • Financial time series data
  • symbolic aggregate approximation (SAX)
  • time series data

ASJC Scopus subject areas

  • Information Systems
  • Software

Cite this

Barnaghi, P. M., Abu Bakar, A., & Ali Othman, Z. (2012). Enhanced symbolic aggregate approximation method for financial time series data representation. In Proceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012 (pp. 790-795). [6528740]

Enhanced symbolic aggregate approximation method for financial time series data representation. / Barnaghi, Peiman Mamani; Abu Bakar, Azuraliza; Ali Othman, Zulaiha.

Proceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012. 2012. p. 790-795 6528740.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Barnaghi, PM, Abu Bakar, A & Ali Othman, Z 2012, Enhanced symbolic aggregate approximation method for financial time series data representation. in Proceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012., 6528740, pp. 790-795, 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012, Taipei, 23/10/12.
Barnaghi PM, Abu Bakar A, Ali Othman Z. Enhanced symbolic aggregate approximation method for financial time series data representation. In Proceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012. 2012. p. 790-795. 6528740
Barnaghi, Peiman Mamani ; Abu Bakar, Azuraliza ; Ali Othman, Zulaiha. / Enhanced symbolic aggregate approximation method for financial time series data representation. Proceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012. 2012. pp. 790-795
@inproceedings{540382843a82410cb820b788e0b1fc62,
title = "Enhanced symbolic aggregate approximation method for financial time series data representation",
abstract = "Data representation is one of the most important tasks in time series data pre-processing. Time series data representation is required to make the data more suitable for data mining specifically for prediction. Time series data is characterized by its numerical and continuous values. One of the data representation methods for time series is the Symbolic Aggregate Approximation (SAX) which uses mean values as the basis of representation of the data. However. representing the time series financial data with the mean value often causes the loss of patterns that can describes important pieces of information. The aim of this study is to propose an enhancement of SAX representation purposely for the financial time series data. The Enhanced SAX (EN-SAX) adds two new values to the original mean value for each segment in SAX. These values enable better representation for each segment in a lower dimension and keep some of the important patterns that are meaningful in financial time series data. The experimental results show that the EN-SAX representation manages to give lower error rates compared to SAX and improves the prediction accuracy.",
keywords = "dimensionality reduction, Financial time series data, symbolic aggregate approximation (SAX), time series data",
author = "Barnaghi, {Peiman Mamani} and {Abu Bakar}, Azuraliza and {Ali Othman}, Zulaiha",
year = "2012",
language = "English",
isbn = "9788994364193",
pages = "790--795",
booktitle = "Proceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012",

}

TY - GEN

T1 - Enhanced symbolic aggregate approximation method for financial time series data representation

AU - Barnaghi, Peiman Mamani

AU - Abu Bakar, Azuraliza

AU - Ali Othman, Zulaiha

PY - 2012

Y1 - 2012

N2 - Data representation is one of the most important tasks in time series data pre-processing. Time series data representation is required to make the data more suitable for data mining specifically for prediction. Time series data is characterized by its numerical and continuous values. One of the data representation methods for time series is the Symbolic Aggregate Approximation (SAX) which uses mean values as the basis of representation of the data. However. representing the time series financial data with the mean value often causes the loss of patterns that can describes important pieces of information. The aim of this study is to propose an enhancement of SAX representation purposely for the financial time series data. The Enhanced SAX (EN-SAX) adds two new values to the original mean value for each segment in SAX. These values enable better representation for each segment in a lower dimension and keep some of the important patterns that are meaningful in financial time series data. The experimental results show that the EN-SAX representation manages to give lower error rates compared to SAX and improves the prediction accuracy.

AB - Data representation is one of the most important tasks in time series data pre-processing. Time series data representation is required to make the data more suitable for data mining specifically for prediction. Time series data is characterized by its numerical and continuous values. One of the data representation methods for time series is the Symbolic Aggregate Approximation (SAX) which uses mean values as the basis of representation of the data. However. representing the time series financial data with the mean value often causes the loss of patterns that can describes important pieces of information. The aim of this study is to propose an enhancement of SAX representation purposely for the financial time series data. The Enhanced SAX (EN-SAX) adds two new values to the original mean value for each segment in SAX. These values enable better representation for each segment in a lower dimension and keep some of the important patterns that are meaningful in financial time series data. The experimental results show that the EN-SAX representation manages to give lower error rates compared to SAX and improves the prediction accuracy.

KW - dimensionality reduction

KW - Financial time series data

KW - symbolic aggregate approximation (SAX)

KW - time series data

UR - http://www.scopus.com/inward/record.url?scp=84880971549&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880971549&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84880971549

SN - 9788994364193

SP - 790

EP - 795

BT - Proceedings - 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (NISS, ICMIA and NASNIT), ISSDM 2012

ER -