首页>
外国专利>
Time series anomaly detection, anomaly classification, and transition analysis using k-nearest neighbor and logistic regression approaches
Time series anomaly detection, anomaly classification, and transition analysis using k-nearest neighbor and logistic regression approaches
展开▼
机译:时间序列异常检测,异常分类和使用k最近邻和逻辑回归方法的转换分析
展开▼
页面导航
摘要
著录项
相似文献
摘要
PROBLEM TO BE SOLVED: To provide a method and a system for time series transition analysis of data. The method is to receive time series data, generate a training data set containing randomized data points, and use a set of randomized data points within a time window. Generating a combination of randomized data points, calculating distance values based on a combination of randomized data points, and generating a classifier based on multiple calculated distance values. Includes using a classifier to determine the probability that new time series data generated during a new run of a process will match the time series data. [Selection diagram] Fig. 2
展开▼