ImputAnom: Anomaly Detection Framework Using Imputation Methods for Univariate Time Series

Tirana Noor Fatyanosa, Mahendra Data, Neni Alya Firdausanti, Putu Hangga Nan Prayoga, Israel Mendonça, Masayoshi Aritsugi

Abstract: Anomaly detection plays a crucial role in various domains such as cybersecurity, fraud detection, and industrial asset condition monitoring. In these fields, identifying abnormal patterns or outliers is paramount for the business they support. This paper presents a new framework that utilizes imputation methods to effectively identify anomalies. To evaluate the performance of the proposed framework, experiments were conducted on different datasets that contain anomalies from different domains. Experimental results demonstrate the effectiveness of the framework in helping to detect anomalies. It provides improvements between 8.17% and 165.21% for all datasets. Experimental results also confirm the effectiveness of the proposed framework and its potential to be applied in real-world scenarios.

External IDs:dblp:conf/iiwas/FatyanosaDFPMA23