import pandas as pd
import sqlite3
# 連接到SQLite數(shù)據(jù)庫
conn = sqlite3.connect('financial_records.db')
# 查詢客戶交易記錄
query = '''
SELECT transaction_id, customer_id, transaction_amount, transaction_date, transaction_type
FROM transactions
'''
df = pd.read_sql_query(query, conn)
# 關(guān)閉數(shù)據(jù)庫連接
conn.close()
# 查看數(shù)據(jù)
print(df.head())

# 數(shù)據(jù)清洗：去除重復記錄
df = df.drop_duplicates()
# 處理缺失值：填充或刪除缺失值
df = df.fillna(method='ffill')
# 標準化：將數(shù)值型特征標準化到相同的尺度
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()
df[['transaction_amount']] = scaler.fit_transform(df[['transaction_amount']])
# 查看預處理后的數(shù)據(jù)
print(df.head())

# 提取特征和標簽
X = df[['customer_id', 'transaction_amount', 'transaction_type']]
y = df['default']
# 將類別特征進行獨熱編碼
X = pd.get_dummies(X, columns=['transaction_type'])
# 查看提取后的特征
print(X.head())

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
# 查看劃分后的數(shù)據(jù)集
print(X_train.shape, X_test.shape)

from sklearn.ensemble import RandomForestClassifier
# 初始化隨機森林模型
model = RandomForestClassifier(n_estimators=100, random_state=42)
# 訓練模型
model.fit(X_train, y_train)
# 查看模型訓練效果
print(model)

from sklearn.metrics import accuracy_score, recall_score, f1_score
# 預測測試集
y_pred = model.predict(X_test)
# 計算評估指標
accuracy = accuracy_score(y_test, y_pred)
recall = recall_score(y_test, y_pred)
f1 = f1_score(y_test, y_pred)
# 輸出評估結(jié)果
print(f'準確率：{accuracy}')
print(f'召回率：{recall}')
print(f'F1分數(shù)：{f1}')

# 假設(shè)已經(jīng)有市場行情數(shù)據(jù)的DataFrame
market_data = pd.read_csv('market_data.csv')
# 處理缺失值
market_data = market_data.fillna(method='ffill')
# 標準化
scaler = StandardScaler()
market_data[['price']] = scaler.fit_transform(market_data[['price']])
# 查看預處理后的數(shù)據(jù)
print(market_data.head())

# 提取特征和標簽
X = market_data[['price', 'volume']]
y = market_data['price'].shift(-1)  # 預測下一個時間點的價格
# 去除空值
X = X[:-1]
y = y.dropna()
# 查看提取后的特征
print(X.head())

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
# 查看劃分后的數(shù)據(jù)集
print(X_train.shape, X_test.shape)

from sklearn.linear_model import LinearRegression
# 初始化線性回歸模型
model = LinearRegression()
# 訓練模型
model.fit(X_train, y_train)
# 查看模型訓練效果
print(model)

from sklearn.metrics import mean_squared_error, mean_absolute_error
# 預測測試集
y_pred = model.predict(X_test)
# 計算評估指標
mse = mean_squared_error(y_test, y_pred)
mae = mean_absolute_error(y_test, y_pred)
# 輸出評估結(jié)果
print(f'均方誤差：{mse}')
print(f'平均絕對誤差：{mae}')

# 假設(shè)已經(jīng)有操作風險數(shù)據(jù)的DataFrame
operation_data = pd.read_csv('operation_data.csv')
# 處理缺失值
operation_data = operation_data.fillna(method='ffill')
# 標準化
scaler = StandardScaler()
operation_data[['amount']] = scaler.fit_transform(operation_data[['amount']])
# 查看預處理后的數(shù)據(jù)
print(operation_data.head())

# 提取特征和標簽
X = operation_data[['amount', 'operation_type', 'time']]
y = operation_data['risk']
# 將類別特征進行獨熱編碼
X = pd.get_dummies(X, columns=['operation_type', 'time'])
# 查看提取后的特征
print(X.head())

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
# 查看劃分后的數(shù)據(jù)集
print(X_train.shape, X_test.shape)

from sklearn.tree import DecisionTreeClassifier
# 初始化決策樹模型
model = DecisionTreeClassifier(random_state=42)
# 訓練模型
model.fit(X_train, y_train)
# 查看模型訓練效果
print(model)

# 預測測試集
y_pred = model.predict(X_test)
# 計算評估指標
accuracy = accuracy_score(y_test, y_pred)
recall = recall_score(y_test, y_pred)
f1 = f1_score(y_test, y_pred)
# 輸出評估結(jié)果
print(f'準確率：{accuracy}')
print(f'召回率：{recall}')
print(f'F1分數(shù)：{f1}')