Applying machine learning in accounting research

Bogaerd, Machteld Van den; Aerts, Walter

doi:10.1016/j.eswa.2011.04.172

Cited by 18 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is a widely adopted technique for such academic research [79], as it is designed "for making replicable and valid inferences from texts (or other meaningful matter) to the contexts of their use" (p. 18), and it consists of a scientific tool that "provides new insights, increases a researcher's understanding of particular phenomena, or informs practical actions" [74] (p. 18). In terms of analysing corporate reports, Wen [80] draws the attention on three main methods that are usually used to analyse texts: individual word-count systems that quantify word frequencies and other text characteristics, human-based content analysis which allows researchers to look more closely at the aspects disclosed in the documents, and computer-aided qualitative data analysis systems that use artificial intelligence to analyse text documents [80], citing [81]. Considering these options, similar to Cosma et al [61], the manual content analysis and interpretation was the chosen method for this study, without the use of any specific software.…”

Section: Methods Used For the Assessment Of The Quality Sdg Score Based On A Content Analysis Approachmentioning

confidence: 99%

Is Financial Information Influencing the Reporting on SDGs? Empirical Evidence from Central and Eastern European Chemical Companies

Nechita¹,

Manea²,

Nichita³

et al. 2020

Sustainability

View full text Add to dashboard Cite

Since the adoption of the sustainable development goals (SDGs) endorsed by United Nations in 2015, which envision an ambitious representation of what the world could look like in the upcoming years, companies have been employing different approaches to identify and report on their contributions to the SDGs, albeit facing various difficulties due to the goals’ novelty, legislative deficiencies, and the complexity of connections between SDGs and business performance. In this context, the relationship between sustainability practices and firms’ financial performance has attracted much attention among researchers. Consequently, this research is focusing on determining to what extent the financial indicators disclosed in the annual reports are impacting the quality of non-financial reporting based on SDGs in the case of chemical companies operating in Central and Eastern European countries (Czech Republic, Hungary, Poland, Romania and Slovakia), within the time frame of 2015–2019. To reach our objective, a score-based content analysis was applied to assess the measure of SDGs achievement, and subsequently, the SDG score was used in multiple linear regression models with several financial data-based independent variables. The main findings highlight the companies’ contribution to the SDGs, particularly to those addressing the environment and decent work conditions for employees, and suggest that the research and development costs and other intangibles represent the most influential variable in explaining the variation in the firms’ SDG score. Academics, businesses and legislative bodies may find these results valuable in their corresponding activities: theoretical, pragmatic or statutory.

show abstract

Section: Methods Used For the Assessment Of The Quality Sdg Score Based On A Content Analysis Approachmentioning

confidence: 99%

Is Financial Information Influencing the Reporting on SDGs? Empirical Evidence from Central and Eastern European Chemical Companies

Nechita¹,

Manea²,

Nichita³

et al. 2020

Sustainability

View full text Add to dashboard Cite

show abstract

“…To identify all relevant news items on a firm and to code them as favourable, unfavourable or balanced according to the tenor of its content, we use a text classification software based on the LPUalgorithm of Liu et al (2002Liu et al ( , 2003. This procedure uses artificial intelligence to classify news items and has been shown to yield high accuracy rates (Liu et al, 2002(Liu et al, , 2003Van den Bogaerd & Aerts, 2011;Zhang & Zuo, 2009). Using artificial intelligence, the algorithm develops its own mode of operation to classify articles according to content type.…”

Section: Media Reputationmentioning

confidence: 99%

Does media reputation affect properties of accounts payable?

Bogaerd

Aerts

2015

European Management Journal

Self Cite

View full text Add to dashboard Cite

“…This has been predicted by several scholars (AICPA 2015; Kokina and Davenport 2017;Sutton, Holt, and Arnold 2016;Vasarhelyi, Sun, and Issa 2016). Numerous articles suggest AI research ideas in this context, present conceptual frameworks, and predict how the impact of AI may happen in auditing (ICAEW 2018;Kokina and Davenport 2017;Omoteso 2012;Raschke et al 2018;Sun 2019;Van den Bogaerd and Aerts 2011;Vasarhelyi, Sun, and Issa 2016). The audit process includes tasks where artificial intelligence and automation can potentially benefit the audit firms and their clients.…”

Section: Artificial Intelligence and Auditingmentioning

confidence: 99%

Artificial intelligence and auditing in small‐ and medium‐sized firms: Expectations and applications

et al. 2022

View full text Add to dashboard Cite

Auditing is a field of expertise often mentioned as being ripe for automation using artificial intelligence methods at all levels of operations. Primarily, the application of artificial intelligence (AI) in the auditing profession is done by and for large organizations, leveraging large datasets. While AI approaches for big data are continually improving, methods for small data are scarce. Yet most firms in the world employ fewer than 50 people and can, therefore, rarely rely on big data for automation. In our study, we ask auditors, who mainly audit SMEs, about their expectations towards the impact of AI on the auditing profession and where they expect it to provide the most value when it comes to auditing SMEs. We find that these auditors expect significant improvements in their own efficiency on the job, that learning to use AI applications will not be a challenge for them, and that the use of AI in auditing firms will become mandatory in the future. They expect the performance of certain tasks to become AI-augmented, including risk assessment of individual transactions, conducting audit interviews, performing all manners of analysis, writing confirmation letters, performing the final verification of annual reports, and performing physical observations. Considering these results, we discuss the potential impact of these developments, such as how AI could make the auditing process more effective and efficient but also how AI could lead to an even higher concentration of the auditing service industry.

show abstract

Applying machine learning in accounting research

Cited by 18 publications

References 23 publications

Is Financial Information Influencing the Reporting on SDGs? Empirical Evidence from Central and Eastern European Chemical Companies

Is Financial Information Influencing the Reporting on SDGs? Empirical Evidence from Central and Eastern European Chemical Companies

Does media reputation affect properties of accounts payable?

Artificial intelligence and auditing in small‐ and medium‐sized firms: Expectations and applications

Contact Info

Product

Resources

About