Accounting Info Systems (SIG ASYS)
Loading...
Paper Type
Complete
Paper Number
1104
Description
To this day, the potential of natural language processing (NLP) is investigated in different domains – auditing is one of them. Internal auditors have to deal with a large number of different business documents, which they have to analyze in various ways. The purpose of this structured literature review is to identify which document types have been addressed in audit-related research and which NLP methods have been used to process these documents. As a result, it was shown that within the main audit areas of compliance, risk identification and especially fraud detection, there is little diversity in the analyzed document types as well as a lack in the use of internal company data. It was also shown that the semantic text level is comparatively rarely taken into account and only a few studies had a direct exchange with auditors from the field to gather requirements or to additionally assess the results.
Recommended Citation
Schumann, Gerrit and Marx Gómez, Jorge, "Natural Language Processing in Internal Auditing – a Structured Literature Review" (2021). AMCIS 2021 Proceedings. 1.
https://aisel.aisnet.org/amcis2021/sig_acctinfosystem_asys/sig_acctinfosystem_asys/1
Natural Language Processing in Internal Auditing – a Structured Literature Review
To this day, the potential of natural language processing (NLP) is investigated in different domains – auditing is one of them. Internal auditors have to deal with a large number of different business documents, which they have to analyze in various ways. The purpose of this structured literature review is to identify which document types have been addressed in audit-related research and which NLP methods have been used to process these documents. As a result, it was shown that within the main audit areas of compliance, risk identification and especially fraud detection, there is little diversity in the analyzed document types as well as a lack in the use of internal company data. It was also shown that the semantic text level is comparatively rarely taken into account and only a few studies had a direct exchange with auditors from the field to gather requirements or to additionally assess the results.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.