Analysis of methods detecting non-technical losses of electricity

The application of data mining in fraud detection has led to advances in detection of fraudulent transactions in banking and e-commerce. At the same time, despite the urgency of the problem of energy thefts, currently, there is no effective solution to detect unmetered consumption, using existing data streams in energy companies. This paper presents a study of the problem of energy thefts, analysis of the available data streams, opportunities and challenges of using these data to reveal the facts of energy thefts.