Abstract:With the rapid development of infrastructure including power grids, managers in power industry these days are faced with an increasingly severe problem of theft of electricity. Theft of electricity has negative effects on many socioeconomic aspects, including impacting stable growth of economic for power enterprises and social development. The traditional anti-theft means require officers checking the integrity of kilowatt-hour meter and the correctness of wiring house by house, which requires enormous manpower and material resources. With the advancement of information collecting technology, power enterprises now possess relatively complete database of power consumption. As a result, performing data mining on existing database and identifying abnormal users has become a hot topic in the field of information technology. The purpose of this paper is to identify abnormal users with machine learning algorithms, providing power enterprises with means to detect theft of electricity at lower cost. The contributions of this paper are listed as follows: 1) Propose various features, providing means to describe users’ behaviors. First, monistic features (mean, difference, coefficient of variation, range, standard deviation), binary features (cosine similarity, Pearson product-moment correlation coefficient) and multivariate features are calculated based on user power consumption (monthly, seasonally, yearly, per holiday, per workday). Then principal component analysis is used to perform dimensionality reduction on the dataset. The dataset is finally scaled and oversampled to form the feature dataset. 2) Study various classification algorithms, optimize hyperparameters, providing models to detect theft of electricity. In this paper, the mechanism behind support vector machine, back-propagation neural network, random forest and XGBoost is studied and the hyperparameters are optimized. 3) Compare and evaluate different classifiers, and provide suggestions for real-world application. In this paper, different classifiers are evaluated with different metrics and the best classifier is recommended based on real-world dataset and different application scenarios.

Mining and Predicting Smart Device User Behavior

Mining Device-Specific Apps Usage Patterns from Appstore Big Data.

A Weekly Load Data Mining Approach Based on Hidden Markov Model

Mining Device-Specific Apps Usage Patterns from Large-Scale Android Users

Establishing Smartphone User Behavior Model Based on Energy Consumption Data

Sensor-Based Human Activity Mining Using Dirichlet Process Mixtures of Directional Statistical Models

Spatio-Temporal Routine Mining on Mobile Phone Data

User Behavior Pattern Analysis And Prediction Based On Mobile Phone Sensors

Theft of Electricity Detection Based on Supervised Learning

Characterizing a User from Large-Scale Smartphone-Sensed Data.

Electrical Behavior Association Mining for Household ShortTerm Energy Consumption Forecasting

Mining Behavioral Patterns from Millions of Android Users

A novel user behavior analysis and prediction algorithm based on mobile social environment

Characterizing and Predicting Individual Traffic Usage of Mobile Application in Cellular Network

Mining consuming Behaviors with Temporal Evolution for Personalized Recommendation in Mobile Marketing Apps

Mining human periodic behaviors via tensor factorization and entropy

Individualized Time-Series Segmentation for Mining Mobile Phone User Behavior

A novel user behavior prediction model based on automatic annotated behavior recognition in smart home systems

Modeling Spatio-Temporal App Usage for a Large User Population.

Mining Long-Term Stealthy User Behaviors on High Speed Links

Pattern-Growth Based Mining Mouse-Interaction Behavior for an Active User Authentication System.