python from sklearn.ensemble import IsolationForest import pandas as pd

假设 df 包含每个账户的特征

features = ['tx_count_7d', 'avg_amt', 'cnt_parties', 'pct_below_reporting_threshold'] model = IsolationForest(contamination=0.01, random_state=42) df['anomaly_score'] = model.fit_predict(df[features])

标记为 -1 的是前 1% 的离群账户