Artificial Neural Network Models for Predicting Patterns in Auditing Monthly Balances
研究人工神经网络模型在审计财务账户时预测和识别模式的能力，通过制造企业的月度余额数据构建模型，发现神经网络有助于分析需要进一步调查的未审计财务数据模式。
Artificial neural networks (ANNs) are a computing paradigm that can be used as a basis for building intelligent information systems. The purpose of this paper is to provide additional evidence as to the ability of an ANN model to forecast and recognise patterns when auditing financial accounts. The present study examines the predictive ability of an ANN by building models using monthly balances of a manufacturing firm. This study uses backpropagation algorithm as learning algorithm. The monthly balances are considered a time-series and the target is to observe the non-linear dynamics and the relationships between accounts. Furthermore, the certain seeded material errors with signals from the ANN model are investigated. The results achieved indicate that neural networks seem promising for analysing patterns that result in a need for additional investigations of unaudited financial data.