AN EVALUATION OF UNSUPERVISED OUTLIER DETECTIONMETHODS FOR UNIVARIATE TIME SERIES DATA IN FINANCIALTRANSACTIONS

Abstract

An essential problem in finance application areas is identifying abnormal subsequences in time series data. Despite the wide range of outlier detection algorithms, no substantial research has been conducted to thoroughly investigate and assess the various methodologies, particularly in the financial industry. This study focuses on comparing and contrasting the outcomes of various unsupervised algorithms. The findings reveal that the Local Outlier Factor technique outperforms the other methods in terms of precision, recall, and F1-score. The research provides valuable insights for financial institutions and businesses looking to improve their identification of abnormalities systems and highlights the importance of choosing the appropriate unsupervised outlier detection method for financial transaction data. The outcomes of this study can be used to inform future research and development in the area of financial
unusual case detection.