Real Estate Transaction Anomaly Detection System

Develop a Python-based anomaly detection system using scikit-learn and pandas that identifies potential fraudulent or irregular real estate transactions. The script must process large Excel/Sheets datasets of property transactions, apply unsupervised machine learning techniques like isolation forests and local outlier factor algorithms to detect statistical irregularities. Create an automated reporting mechanism that flags suspicious transactions with detailed contextual analysis and generates compliance reports exportable to Google Sheets.