2019
DOI: 10.1111/acfi.12584
|View full text |Cite
|
Sign up to set email alerts
|

A machine learning analysis of citation impact among selected Pacific Basin journals

Abstract: This study uses a machine learning approach to identify and predict factors which influence citation impacts across five Pacific Basin journals: Abacus, Accounting & Finance, Australian Journal of Management, Australian Accounting Review and the Pacific Accounting Review from 2008 to 2018. The machine learning results indicate that citation impact is mostly influenced by: length of a journal article; the field of research (particularly environmental accounting), sample size; whether the sample is local or inte… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
4
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
7

Relationship

0
7

Authors

Journals

citations
Cited by 11 publications
(4 citation statements)
references
References 169 publications
0
4
0
Order By: Relevance
“…To test the validity and reliability of this result, we applied several other types of analysis suggested by researchers working with literature reviews. For example, Dumay and Cai (2014) and Jones and Alam (2019) argue that citation impact factors are increasingly important because they identify the most influential articles. Highly cited articles represent a “corpus of scholarly literature” that can help “develop insights, critical reflections, future research paths and research questions” (Massaro et al , 2016, p. 767).…”
Section: Methodsmentioning
confidence: 99%
“…To test the validity and reliability of this result, we applied several other types of analysis suggested by researchers working with literature reviews. For example, Dumay and Cai (2014) and Jones and Alam (2019) argue that citation impact factors are increasingly important because they identify the most influential articles. Highly cited articles represent a “corpus of scholarly literature” that can help “develop insights, critical reflections, future research paths and research questions” (Massaro et al , 2016, p. 767).…”
Section: Methodsmentioning
confidence: 99%
“…The gradient-boosting algorithm (Friedman, 2001(Friedman, , 2002Friedman et al, 2000) implements a gradient descent algorithm to the boosting model, significantly improving its efficiency for applications. In accounting and finance research, ensemble learning algorithms have been used, for example, to predict financial failure (Jiang & Jones, 2018;Jones, 2017), to identify and predict features that influence citation impacts (Jones & Alam, 2019), to detect fraud (Bao et al, 2020;Gepp et al, 2021), to model the conditional risk premium (Hoang & Faff, 2021), to improve managerial estimates in financial reports (Ding et al, 2020), to predict corporate bond default (Lu & Zhuo, 2021), and to examine the value and relative importance of various board and firm characteristics in predicting workplace diversity (Ranta & Ylinen, 2023). The gradient boosting algorithm was introduced in the works of Friedman et al (2000), Friedman (2001), and Friedman and Popescu (2003).…”
Section: A Tree-based Gradient-boosting Modelmentioning
confidence: 99%
“…Given that such methods are still only emerging, their applications in the fields of accounting or finance are limited; however, some initial literature reviews are emerging ( e.g. Cai et al, 2019;Jones and Alam, 2019). However, we are able to draw upon developments in fields of machine learning techniques which are increasingly focussed not just on static word frequency measures but also on topic detection algorithms (Marrone, 2020).…”
Section: Note(s)mentioning
confidence: 99%