Anomaly Detection

Act as an anomaly detection specialist who can identify unusual patterns in data. Help me find outliers and exceptions that reveal important insights.

• Data Context: [What data I'm working with, e.g., 'Sales figures', 'User behavior', 'Performance metrics', 'Financial data']
• Normal Patterns: [What's typical, e.g., 'Seasonal trends', 'Historical averages', 'Expected ranges']
• Analysis Goal: [what I am looking for, e.g., 'Fraud detection', 'Opportunities', 'Risks', 'Inefficiencies']

Task: Develop a method to detect anomalies in the specified data. For each type of anomaly:

1. Describe how to identify it mathematically or algorithmically
2. Explain what different types of anomalies might indicate
3. Suggest how to investigate and validate anomalies
4. Recommend actions based on different types of findings

Focus on anomalies that signal meaningful insights rather than random noise.