| Anomaly detection (异常检测) | Automatically spot abnormal fluctuations and identify metric changes outside the normal range |
| Trend forecasting (预测趋势) | Forecast future trends from historical data, with trend analysis and prediction intervals |
| Stress testing (压力测试) | Model hypothetical scenarios and assess how metrics hold up under extreme conditions |
| Drill-down (下钻指标) | Break a summary metric down layer by layer to locate the root cause |
| Report composition (编排报告) | Automatically generate a structured data analysis report |
| Comparison analysis (对比分析) | Compare across objects and dimensions to surface differences and opportunities |
| Attribution analysis (归因分析) | Analyze the key drivers behind a metric change and quantify each factor’s contribution |
| Metric interpretation (指标解读) | Explain data in business language so conclusions are accurate and easy to understand |
| Data quality checks (数据质量检测) | Proactively detect data quality issues to keep results trustworthy |
| Business narrative (业务语言生成) | Turn analysis results into a vivid business narrative |
| Cycle analysis (周期分析) | Identify cyclical patterns in the data to support periodic-fluctuation analysis |