Data mining as a tool for detecting anomalies and patterns in the internal audit processes of medium-sized Venezuelan enterprises

Abstract

This study aims to analyze how data mining can be applied to detect anomalies and patterns within the internal audit processes of medium-sized Venezuelan enterprises. The research is descriptive in nature and follows a qualitative, phenomenological–interpretive approach. Data were collected through semi-structured interviews with experts in internal audit departments from selected Venezuelan entities, all of whom possessed specialized knowledge in data mining techniques. The results indicate that data mining optimizes internal audit processes by enabling the automation of operational tasks, reducing the time required to detect irregularities, and strengthening risk management. The study also found that implementing continuous auditing facilitates the early detection of deviations while significantly reducing the operational burden on auditors. Data mining's capacity to analyze large volumes of data and uncover hidden patterns aligns with established theoretical evidence. However, successful implementation in the Venezuelan context faces organizational challenges, including resistance to change, financial limitations, and a gap in specialized technical skills. The findings highlight the urgent need for organizations to invest in both technology and specialized training in advanced analytical tools—such as Power BI, SQL, and ACL—to ensure a more effective and modern internal audit framework.