Anomaly Detection Techniques in Smart City: A Review from a Framework Perspective

Abstract

A smart city is a notion in which a city may effectively and efficiently manage its resources in order to improve the quality of life of its residents. With the existence of information technology as an enabler, the city is able to carry out sensing, understanding and acting at a certain level. Anomaly detection is a growing field of research in almost all domains, one of which is smart cities. In a smart city, anomaly detection enters into the context of understanding, where the data received, both cyber physical system sensors and user generated content are processed to obtain unusual data that can affect predictions or forecasts. Many studies have been carried out related to anomaly detection methods and techniques related to the smart city environment but have not been categorized according to the domain or framework that is in accordance with the smart city model. This study aims to present the results of a systematic literature review on anomaly detection techniques categorized according to the Garuda Smart City Model, a model designed for smart city development in Indonesia. Based on the results of a systematic literature review, many research gaps have been found for anomaly detection in a smart city environment related to performance limitations, the effect of implementation methods, and the use of anomaly detection as part of the prediction and forecasting methods.