This study explores the application of deep learning techniques, particularly the Autoencoder algorithm, in credit card fraud detection to enhance the performance and reliability of anomaly detection methods in financial transactions. The research begins with the Isolation Forest algorithm, achieving a detection accuracy of 26% for the top 1000 transactions. Subsequently, the Autoencoder algorithm, an unsupervised deep neural network model, improves the detection accuracy to 33.6% in the best case, despite some fluctuations. The study highlights the strong feature extraction capability and adaptability of deep learning models, suggesting their potential to surpass traditional methods. However, the dataset's high imbalance, with only 0.17% of transactions being fraudulent, poses a significant challenge. The findings emphasize the need for further experimentation and optimization of network structures and hyperparameters to achieve more stable and efficient fraud detection. The research provides valuable insights and references for future studies in financial fraud detection using deep learning methodologies.This study explores the application of deep learning techniques, particularly the Autoencoder algorithm, in credit card fraud detection to enhance the performance and reliability of anomaly detection methods in financial transactions. The research begins with the Isolation Forest algorithm, achieving a detection accuracy of 26% for the top 1000 transactions. Subsequently, the Autoencoder algorithm, an unsupervised deep neural network model, improves the detection accuracy to 33.6% in the best case, despite some fluctuations. The study highlights the strong feature extraction capability and adaptability of deep learning models, suggesting their potential to surpass traditional methods. However, the dataset's high imbalance, with only 0.17% of transactions being fraudulent, poses a significant challenge. The findings emphasize the need for further experimentation and optimization of network structures and hyperparameters to achieve more stable and efficient fraud detection. The research provides valuable insights and references for future studies in financial fraud detection using deep learning methodologies.