Sentiment analysis in decision sciences research: An illustration to IT governance

Qing Cao, Mark Andrew Thompson, Yang Yu

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

We propose a viable procedure to mine opinions from a large collection of texts. We demonstrate how this procedure can be applied in business research by constructing IT governance measures. Specifically, we generate a set of IT governance measures using a multi-label classification method and compare against our proposed sentiment analysis procedure. Using 10-K forms to develop our measures of IT governance, we examine the role of IT governance on firm performance. We find evidence of the five dimensions of IT governance (as proposed by the IT Governance Institute), but only one dimension significantly explains firm performance (i.e., strategic alignment) using the multi-label classification method. Using our new procedure to construct IT governance measures, we find significant evidence of four dimensions in explaining firm performance.

Original languageEnglish (US)
Pages (from-to)1010-1015
Number of pages6
JournalDecision Support Systems
Volume54
Issue number2
DOIs
StatePublished - Jan 1 2013
Externally publishedYes

Fingerprint

Decision Support Techniques
Labels
Research
Industry
IT governance
Decision science
Sentiment analysis
Sentiment
Governance
Firm performance

Keywords

  • IT governance
  • Sentiment analysis
  • Text mining

ASJC Scopus subject areas

  • Management Information Systems
  • Information Systems
  • Developmental and Educational Psychology
  • Arts and Humanities (miscellaneous)
  • Information Systems and Management

Cite this

Sentiment analysis in decision sciences research : An illustration to IT governance. / Cao, Qing; Thompson, Mark Andrew; Yu, Yang.

In: Decision Support Systems, Vol. 54, No. 2, 01.01.2013, p. 1010-1015.

Research output: Contribution to journalArticle

@article{7088a543e8d84dcab05d4347ab19561c,
title = "Sentiment analysis in decision sciences research: An illustration to IT governance",
abstract = "We propose a viable procedure to mine opinions from a large collection of texts. We demonstrate how this procedure can be applied in business research by constructing IT governance measures. Specifically, we generate a set of IT governance measures using a multi-label classification method and compare against our proposed sentiment analysis procedure. Using 10-K forms to develop our measures of IT governance, we examine the role of IT governance on firm performance. We find evidence of the five dimensions of IT governance (as proposed by the IT Governance Institute), but only one dimension significantly explains firm performance (i.e., strategic alignment) using the multi-label classification method. Using our new procedure to construct IT governance measures, we find significant evidence of four dimensions in explaining firm performance.",
keywords = "IT governance, Sentiment analysis, Text mining",
author = "Qing Cao and Thompson, {Mark Andrew} and Yang Yu",
year = "2013",
month = "1",
day = "1",
doi = "10.1016/j.dss.2012.10.026",
language = "English (US)",
volume = "54",
pages = "1010--1015",
journal = "Decision Support Systems",
issn = "0167-9236",
publisher = "Elsevier",
number = "2",

}

TY - JOUR

T1 - Sentiment analysis in decision sciences research

T2 - An illustration to IT governance

AU - Cao, Qing

AU - Thompson, Mark Andrew

AU - Yu, Yang

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We propose a viable procedure to mine opinions from a large collection of texts. We demonstrate how this procedure can be applied in business research by constructing IT governance measures. Specifically, we generate a set of IT governance measures using a multi-label classification method and compare against our proposed sentiment analysis procedure. Using 10-K forms to develop our measures of IT governance, we examine the role of IT governance on firm performance. We find evidence of the five dimensions of IT governance (as proposed by the IT Governance Institute), but only one dimension significantly explains firm performance (i.e., strategic alignment) using the multi-label classification method. Using our new procedure to construct IT governance measures, we find significant evidence of four dimensions in explaining firm performance.

AB - We propose a viable procedure to mine opinions from a large collection of texts. We demonstrate how this procedure can be applied in business research by constructing IT governance measures. Specifically, we generate a set of IT governance measures using a multi-label classification method and compare against our proposed sentiment analysis procedure. Using 10-K forms to develop our measures of IT governance, we examine the role of IT governance on firm performance. We find evidence of the five dimensions of IT governance (as proposed by the IT Governance Institute), but only one dimension significantly explains firm performance (i.e., strategic alignment) using the multi-label classification method. Using our new procedure to construct IT governance measures, we find significant evidence of four dimensions in explaining firm performance.

KW - IT governance

KW - Sentiment analysis

KW - Text mining

UR - http://www.scopus.com/inward/record.url?scp=84871720559&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871720559&partnerID=8YFLogxK

U2 - 10.1016/j.dss.2012.10.026

DO - 10.1016/j.dss.2012.10.026

M3 - Article

AN - SCOPUS:84871720559

VL - 54

SP - 1010

EP - 1015

JO - Decision Support Systems

JF - Decision Support Systems

SN - 0167-9236

IS - 2

ER -