Accelerating Financial Statement Audit Procedures: an Examination of Machine Learning for Risk Assessment and Anomaly Detection

EasyChair Preprint 14261

Abstract

The increasing volume and complexity of financial data pose significant challenges for traditional financial statement audit procedures, often leading to increased audit time and costs. This research investigates the potential of machine learning to accelerate and enhance the efficiency of financial statement audits, focusing on two key areas: risk assessment and anomaly detection. By leveraging ML's ability to analyze vast datasets and identify patterns, this study explores how auditors can make more informed risk assessments and efficiently detect potential misstatements. The research will examine various ML algorithms, including supervised, unsupervised, and semi-supervised learning techniques, to evaluate their effectiveness in identifying financial anomalies and predicting audit risk factors. Furthermore, the study will address the challenges and limitations of implementing ML in audit procedures, such as data quality, model interpretability, and ethical considerations. This research aims to provide practical insights for audit professionals and contribute to the ongoing dialogue on leveraging advanced technologies to improve audit quality and efficiency in the face of evolving financial reporting landscapes.

Keyphrases: Deep Neural Networks (DNNs), Machine Learning (ML), compress convolutional neural networks (CNNs)

@booklet{EasyChair:14261,
  author    = {Abill Robert},
  title     = {Accelerating Financial Statement Audit Procedures: an Examination of Machine Learning for Risk Assessment and Anomaly Detection},
  howpublished = {EasyChair Preprint 14261},
  year      = {EasyChair, 2024}}