Unsupervised BLSTM-Based Electricity Theft Detection with Training Data Contaminated.
Qiushi LiangShengjie ZhaoJiangfan ZhangHao DengPublished in: ACM Trans. Cyber Phys. Syst. (2024)
Keyphrases
- training data
- supervised learning
- noisy data
- data sets
- detection rate
- detection accuracy
- learning algorithm
- object detection
- short term
- test data
- training examples
- unsupervised learning
- automatic detection
- training set
- decision trees
- training dataset
- electric power
- detection method
- domain knowledge
- prior knowledge
- information security
- training samples
- anomaly detection
- support vector machine
- active learning