Log Service - Added Machine Learning Functions in Log Service
Target customers: developers, administrators, data analysts. Features released: Support for diverse algorithms and multiple calling methods. You are allowed to call machine learning algorithms using SELECT statements and machine learning functions to analyze the characteristics of a field or several fields over a period of time in Log Service. Especially for the scenarios related to analyzing time series data, Log Service provides abundant time series analysis algorithms. This helps you to address scenarios, such as time series prediction, time series anomaly detection, sequence decomposition, and multiple time series clustering. These algorithms are compatible with SQL standard APIs. This allows you to easily use these algorithms to improve the efficiency of analyzing and solving issues. Advantages:1. Support for smooth operations for time series data.2. Support for multiple time series algorithms, such as prediction, anomaly detection, change point detection, breakpoint detection, and multiple-period estimation.3. Support for decomposition operations for time series data.4. Support for multiple clustering algorithms for time series data.5. Support for pattern mining for multiple attributes, such as the number column and text column.