Upload 5 files

app.py

@@ -4,19 +4,29 @@ import pandas as pd
 import joblib
 from sklearn.cluster import HDBSCAN
 import numpy as np
-import logger_service
+import logger_service  # 🌟 保持：匯入外部日誌服務模組
 
-# --- 1. 載入模型與歷史資料 (實現動態 Rule 的關鍵) ---
+# 初始化歷史日誌串列
+log_history = []
+
+# --- 1. 載入模型與歷史資料 ---
 try:
     le_action = joblib.load('le_action.pkl')
     le_ip = joblib.load('le_ip.pkl')
     
-    # 🌟 載入歷史資料作為 HDBSCAN 即時比對的基準
     baseline_df = pd.read_csv('baseline_logs.csv')
-    baseline_df['hour'] = pd.to_datetime(baseline_df['timestamp']).dt.hour
+    baseline_df['timestamp'] = pd.to_datetime(baseline_df['timestamp'])
+    
+    # 🌟 保持：預載歷史日誌顯示
+    log_history = baseline_df.sort_values(by='timestamp', ascending=False).head(20).to_dict('records')
+    
+    baseline_df['month'] = baseline_df['timestamp'].dt.month
+    baseline_df['day'] = baseline_df['timestamp'].dt.day
+    baseline_df['hour'] = baseline_df['timestamp'].dt.hour
     
-    # 預先將歷史資料轉成特徵矩陣
     baseline_X = np.column_stack((
+        baseline_df['month'].values,
+        baseline_df['day'].values,
         baseline_df['hour'].values,
         le_ip.transform(baseline_df['ip_address'].values),
         le_action.transform(baseline_df['action'].values)
@@ -26,117 +36,194 @@ try:
     print("✅ AI 模組與歷史基準資料載入成功！")
 except Exception as e:
     AI_READY = False
-    print(f"⚠️ AI 模組載入失敗，請確認 .pkl 與 .csv 檔案是否已上傳。錯誤: {e}")
+    print(f"⚠️ AI 模組載入失敗，請確認檔案。錯誤: {e}")
 
 USER_DB = {
     "student": ["1234", "學生", "同學"], 
-    "teacher": ["1234", "老師", "教授"],
-    "admin": ["1234", "行政", "行政人員"],
-    "ta": ["1234", "TA", "助教"],
-    "sysadmin": ["1234", "系統管理員", "管理員"]
+    "admin": ["1234", "管理", "管理員"],
 }
 
-log_history = baseline_df.sort_values(by='timestamp', ascending=False).head(20).to_dict('records')
-
-# --- 3. 核心邏輯與真實 HDBSCAN 異常偵測 ---
-def check_anomaly(ip, action, log_time):
-    if not AI_READY:
-        return "success" 
-        
-    if action in ["login_attempt", "logout"]:
-        return "success"
+# --- 2. 核心邏輯與真實 HDBSCAN 異常偵測 ---
+def check_anomaly(ip, action, log_time, overall_w, time_w, ip_w, dev_w):
+    if not AI_READY: return "success" 
+    if action in ["login_attempt", "logout"]: return "success"
 
     try:
-        # 1. 將新進來的行為轉為代碼
-        ip_code = le_ip.transform([ip])[0]
-        action_code = le_action.transform([action])[0]
+        ow = float(overall_w)
+        tw = float(time_w)
+        iw = float(ip_w)
+        dw = float(dev_w)
+        
+        clean_ip = str(ip).split(" ")[0] 
+        ip_code = float(le_ip.transform([clean_ip])[0])
+        action_code = float(le_action.transform([action])[0])
         
-        # 2. 擷取當下行為的時間 (Hour)
-        current_hour = pd.to_datetime(log_time).hour
+        current_time = pd.to_datetime(str(log_time))
+        current_month = float(current_time.month)
+        current_day = float(current_time.day)
+        current_hour = float(current_time.hour)
         
-        # 3. 組合這筆新資料的特徵向量
-        new_data_point = np.array([[current_hour, ip_code, action_code]])
+        time_multiplier = tw / 100.0
+        action_scale = (dw / 100.0) * 100.0 
+        
+        scaled_baseline = baseline_X.copy().astype(float)
+        scaled_baseline[:, 0] *= time_multiplier * 10.0  
+        scaled_baseline[:, 1] *= time_multiplier * 0.5   
+        scaled_baseline[:, 2] *= time_multiplier * 2.0   
+        scaled_baseline[:, 3] *= (iw / 100.0) * 10.0     
+        scaled_baseline[:, 4] *= action_scale            
+        
+        v_month = current_month * time_multiplier * 10.0
+        v_day = current_day * time_multiplier * 0.5
+        v_hour = current_hour * time_multiplier * 2.0
+        v_ip = ip_code * (iw / 100.0) * 10.0
+        v_action = action_code * action_scale
+        
+        new_data_point = np.array([[v_month, v_day, v_hour, v_ip, v_action]], dtype=float)
+        
+        # 🌟 智能特徵抽樣 (Smart Sampling) 核心邏輯
+        # 1. 找出歷史資料中，跟「當下準備執行的行為」完全一樣的紀錄 (保留季節性與週期性記憶)
+        same_action_mask = (baseline_X[:, 4] == action_code)
+        relevant_history = scaled_baseline[same_action_mask]
+        
+        # 2. 隨機抽取一部分其他的背景資料 (維持模型對「一般常規作息」的認知)
+        other_history = scaled_baseline[~same_action_mask]
+        if len(other_history) > 800:
+            # 固定隨機種子，確保相同情境下的抽樣穩定度
+            np.random.seed(int(current_day + current_hour)) 
+            indices = np.random.choice(len(other_history), 800, replace=False)
+            other_history = other_history[indices]
+            
+        # 3. 組合出兼顧「運算速度」與「全域歷史視野」的精華資料集
+        fit_X = np.vstack((other_history, relevant_history, new_data_point))
         
-        # 4. 🌟 將新資料與歷史資料合併 (這就是流程圖中的「即時丟入 AI 進行判斷」)
-        combined_X = np.vstack((baseline_X, new_data_point))
+        strictness = max(2, int(2 + (ow / 100.0) * 4))
         
-        # 5. 執行 HDBSCAN 分群
-        model = HDBSCAN(min_cluster_size=5, min_samples=3)
-        clusters = model.fit_predict(combined_X)
+        model = HDBSCAN(
+            min_cluster_size=strictness, 
+            min_samples=2
+        )
         
-        # 6. 檢查最後一筆 (也就是剛剛送進來的新動作) 是否被分類為 -1 (雜訊/異常)
-        if clusters[-1] == -1:
+        # 改用濃縮後的 fit_X 進行快速分群
+        clusters = model.fit_predict(fit_X)
+        
+        if clusters[-1] == -1: 
             return "⚠️異常_作息或行為不符"
-        else:
-            return "success"
+
+        # 距離計算邏輯保持不變
+        if np.any(same_action_mask):
+            history_points = scaled_baseline[same_action_mask]
+            distances = np.linalg.norm(history_points - new_data_point[0], axis=1)
+            min_distance = np.min(distances)
+            
+            max_tolerance_distance = time_multiplier * 15.0 + 1.0
             
+            if min_distance > max_tolerance_distance:
+                return "⚠️異常_作息或行為不符"
+                
+        return "success"
     except ValueError:
-        # 如果遇到沒看過的 IP 或動作，連算都不用算，絕對是異常
-        return "⚠️異常_未知特徵"
-
-def record_log(ip, cookie, account, role, action, status, time_mode, custom_time):
-    global baseline_X, log_history
-    # 決定紀錄的時間
-    if time_mode == "自訂時間 (模擬過去/未來)" and custom_time.strip() != "":
-        log_time = custom_time
+        return "⚠️異常_未知特徵 (出現未授權的 IP 或操作)"
+    except Exception as e:
+        return f"⚠️系統錯誤: {str(e)}"
+
+def record_log(ip, cookie, account, role, action, status, time_mode, custom_time, ow, tw, iw, dw):
+    global baseline_X 
+    if time_mode == "自訂時間 (模擬過去/未來)" and custom_time and str(custom_time).strip() != "":
+        log_time = str(custom_time)
     else:
         tw_tz = datetime.timezone(datetime.timedelta(hours=8))
         log_time = datetime.datetime.now(tw_tz).strftime("%Y-%m-%d %H:%M:%S")
 
-    # 執行 AI 檢查 (現在把 log_time 也傳進去了)
     if status == "success":
-        ai_judgment = check_anomaly(ip, action, log_time)
-        if ai_judgment != "success":
-            status = ai_judgment
+        ai_judgment = check_anomaly(ip, action, log_time, ow, tw, iw, dw)
+        if ai_judgment != "success": status = ai_judgment
 
     new_log = {
-        "timestamp": log_time,
-        "ip_address": ip,
-        "cookie_id": cookie,
-        "account": account,
-        "role": role,
-        "action": action,
-        "status": status
+        "timestamp": log_time, "ip_address": ip, "cookie_id": cookie, 
+        "account": account, "role": role, "action": action, "status": status
     }
-
+    
+    # 🌟 保持：整合外部日誌服務
     try:
-        # 將新行為轉為特徵數值
-        new_hour = pd.to_datetime(log_time).hour
-        new_ip_code = le_ip.transform([ip])[0]
-        new_action_code = le_action.transform([action])[0]
-        new_feature = np.array([[new_hour, new_ip_code, new_action_code]])
-        
-        # 🌟 把新特徵壓入全域矩陣中
-        baseline_X = np.vstack((baseline_X, new_feature))
+        logger_service.push_new_log(new_log)
     except Exception as e:
-        print(f"⚠️ AI 矩陣更新跳過 (可能是未知特徵): {e}")
-
-    # 呼叫外部 Service 進行存檔與上傳
-    logger_service.push_new_log(new_log)
+        print(f"⚠️ 外部日誌推送失敗: {e}")
 
     log_history.insert(0, new_log)
+
+    if status == "success" and AI_READY:
+        try:
+            pd.DataFrame([new_log]).to_csv('baseline_logs.csv', mode='a', header=False, index=False, encoding='utf-8-sig')
+            
+            current_time = pd.to_datetime(str(log_time))
+            clean_ip = str(ip).split(" ")[0]
+            new_row = np.array([[
+                float(current_time.month), 
+                float(current_time.day), 
+                float(current_time.hour), 
+                float(le_ip.transform([clean_ip])[0]), 
+                float(le_action.transform([action])[0])
+            ]], dtype=float)
+            baseline_X = np.vstack((baseline_X, new_row))
+        except Exception as e:
+            pass 
+
     return pd.DataFrame(log_history)
 
-def process_login(ip, cookie, account, password, captcha, time_mode, custom_time):
+# --- 3. 視角切換與登入控制邏輯 ---
+def fast_login(target_account, ip, cookie, time_mode, custom_time, ow, tw, iw, dw):
+    role, name = USER_DB[target_account][1], USER_DB[target_account][2]
+    df = record_log(ip, cookie, target_account, role, "login_attempt", "success", time_mode, custom_time, ow, tw, iw, dw)
+    welcome_msg = f"**{name}** 歡迎您  |  線上人數: 938 (⚡快速登入模式)"
+    target_tab = "student_tab" if role == "學生" else "admin_tab"
+    return gr.update(visible=False), gr.update(visible=True), gr.update(selected=target_tab), "", welcome_msg, df, target_account
+
+def process_login(ip, cookie, account, password, captcha, time_mode, custom_time, ow, tw, iw, dw):
     valid_captcha = "FEIK"
-    if account not in USER_DB: return gr.update(visible=True), gr.update(visible=False), "帳號不存在！", gr.update(), record_log(ip, cookie, account, "Unknown", "login_attempt", "failed_no_account", time_mode, custom_time)
+    def fail_response(msg, df):
+        return gr.update(visible=True), gr.update(visible=False), gr.update(), msg, gr.update(), df
+
+    if account not in USER_DB: 
+        df = record_log(ip, cookie, account, "Unknown", "login_attempt", "failed_no_account", time_mode, custom_time, ow, tw, iw, dw)
+        return fail_response("帳號不存在！", df)
+        
     role, name = USER_DB[account][1], USER_DB[account][2]
-    if captcha.upper() != valid_captcha: return gr.update(visible=True), gr.update(visible=False), "驗證碼錯誤！", gr.update(), record_log(ip, cookie, account, role, "login_attempt", "failed_captcha", time_mode, custom_time)
+    
+    if captcha.upper() != valid_captcha: 
+        df = record_log(ip, cookie, account, role, "login_attempt", "failed_captcha", time_mode, custom_time, ow, tw, iw, dw)
+        return fail_response("驗證碼錯誤！", df)
+        
     if USER_DB[account][0] == password:
         welcome_msg = f"**{name}** 歡迎您  |  線上人數: 938"
-        df = record_log(ip, cookie, account, role, "login_attempt", "success", time_mode, custom_time)
-        return gr.update(visible=False), gr.update(visible=True), "", welcome_msg, df
-    return gr.update(visible=True), gr.update(visible=False), "密碼錯誤！", gr.update(), record_log(ip, cookie, account, role, "login_attempt", "failed_password", time_mode, custom_time)
+        df = record_log(ip, cookie, account, role, "login_attempt", "success", time_mode, custom_time, ow, tw, iw, dw)
+        target_tab = "student_tab" if role == "學生" else "admin_tab"
+        return gr.update(visible=False), gr.update(visible=True), gr.update(selected=target_tab), "", welcome_msg, df
+        
+    df = record_log(ip, cookie, account, role, "login_attempt", "failed_password", time_mode, custom_time, ow, tw, iw, dw)
+    return fail_response("密碼錯誤！", df)
 
-def simulate_click(ip, cookie, account, button_name, time_mode, custom_time):
-    role = USER_DB.get(account, ["", "Unknown"])[1]
-    return record_log(ip, cookie, account, role, button_name, "success", time_mode, custom_time)
+def toggle_role(current_account, ip, cookie, time_mode, custom_time, ow, tw, iw, dw):
+    target_account = "student" if current_account == "admin" else "admin"
+    role, name = USER_DB[target_account][1], USER_DB[target_account][2]
+    df = record_log(ip, cookie, target_account, role, "login_attempt", "success", time_mode, custom_time, ow, tw, iw, dw)
+    welcome_msg = f"**{name}** 歡迎您  |  線上人數: 938 (⚡單鍵切換模式)"
+    target_tab = "student_tab" if role == "學生" else "admin_tab"
+    return gr.update(selected=target_tab), welcome_msg, df, target_account
 
-def logout(ip, cookie, account, time_mode, custom_time):
+def logout(ip, cookie, account, time_mode, custom_time, ow, tw, iw, dw):
     role = USER_DB.get(account, ["", "Unknown"])[1]
-    df = record_log(ip, cookie, account, role, "logout", "success", time_mode, custom_time)
+    df = record_log(ip, cookie, account, role, "logout", "success", time_mode, custom_time, ow, tw, iw, dw)
     return gr.update(visible=True), gr.update(visible=False), "", "", "", "", df
 
+def simulate_click(ip, cookie, account, button_name, time_mode, custom_time, ow, tw, iw, dw):
+    role = USER_DB.get(account, ["", "Unknown"])[1]
+    return record_log(ip, cookie, account, role, button_name, "success", time_mode, custom_time, ow, tw, iw, dw)
+
+def test_anomaly(ip, cookie, account, action_name, test_time_str, ow, tw, iw, dw):
+    role = USER_DB.get(account, ["", "Unknown"])[1]
+    return record_log(ip, cookie, account, role, action_name, "success", "自訂時間 (模擬過去/未來)", test_time_str, ow, tw, iw, dw)
+
 def toggle_time_input(mode):
     return gr.update(visible=(mode == "自訂時間 (模擬過去/未來)"))
 
@@ -144,8 +231,10 @@ def toggle_time_input(mode):
 custom_css = """
 .mock-panel { background-color: var(--background-fill-secondary); padding: 15px; border-radius: 8px; border: 1px dashed var(--border-color-primary); }
 .login-container { max-width: 500px; margin: 0 auto; padding: 20px; }
-.placeholder-box { background-color: var(--background-fill-secondary); border: 2px dashed var(--border-color-primary); text-align: center; color: var(--body-text-color); padding: 40px 0; margin-bottom: 15px; border-radius: 5px; }
 .ntut-title { color: #d9534f; font-size: 22px; font-weight: bold; text-align: center; margin-bottom: 20px;}
+.weight-card { padding: 10px; background: white; border-radius: 8px; box-shadow: 0 1px 3px rgba(0,0,0,0.1); margin-top: 10px; }
+.headless-tabs > div:first-child { display: none !important; }
+.headless-tabs { border: none !important; background: transparent !important; }
 """
 
 with gr.Blocks(title="模擬校園入口與資安防護系統", css=custom_css) as demo:
@@ -159,43 +248,156 @@ with gr.Blocks(title="模擬校園入口與資安防護系統", css=custom_css)
             custom_time_input = gr.Textbox(label="輸入自訂時間", placeholder="格式: YYYY-MM-DD HH:MM:SS", value="2026-03-17 03:00:00", visible=False)
             gr.Markdown("---")
             gr.Markdown("### 📊 即時 Log 與 AI 判定區")
-            log_display = gr.Dataframe(headers=["timestamp", "ip_address", "cookie_id", "account", "role", "action", "status"], datatype=["str", "str", "str", "str", "str", "str", "str"], interactive=False, wrap=True)
+            log_display = gr.Dataframe(value=log_history, headers=["timestamp", "ip_address", "cookie_id", "account", "role", "action", "status"], datatype=["str", "str", "str", "str", "str", "str", "str"], interactive=False, wrap=True)
 
         with gr.Column(scale=2):
             with gr.Column(visible=True, elem_classes="login-container") as login_view:
-                gr.HTML("<div class='placeholder-box'>預留位置：台北科大 Logo 與 e化校園圖片</div>")
                 gr.Markdown("<div class='ntut-title'>校園入口網站 Taipei Tech Portal</div>")
                 login_msg = gr.Markdown(value="", visible=True)
-                with gr.Row(): acc_input = gr.Textbox(label="使用者帳號 (Account)", placeholder="例如: student, admin...")
-                with gr.Row(): pwd_input = gr.Textbox(label="使用者密碼 (Password)", type="password", placeholder="密碼皆為 1234")
+                with gr.Row(): acc_input = gr.Textbox(label="使用者帳號 (Account)", placeholder="student or admin")
+                with gr.Row(): pwd_input = gr.Textbox(label="使用者密碼 (Password)", type="password", placeholder="password : 1234")
                 with gr.Row(): gr.HTML("<div style='background:var(--background-fill-secondary); color:var(--body-text-color); padding:10px; font-weight:bold; letter-spacing: 3px; border:1px solid var(--border-color-primary); text-align:center;'>F E I K</div>")
                 with gr.Row(): captcha_input = gr.Textbox(label="請輸入驗證碼 (Keyin Code)")
                 login_btn = gr.Button("登入 Login", variant="primary")
+                
+                gr.Markdown("---")
+                gr.Markdown("<div style='text-align:center; color:gray; font-size: 0.9em;'>🛠️ 開發與測試專用捷徑</div>")
+                with gr.Row():
+                    btn_fast_student = gr.Button("🚀 快速登入 (學生)", variant="secondary")
+                    btn_fast_admin = gr.Button("🚀 快速登入 (管理員)", variant="secondary")
 
             with gr.Column(visible=False) as dashboard_view:
-                gr.HTML("<div class='placeholder-box'>預留位置：學生學習歷程入口網站橫幅</div>")
                 with gr.Row():
                     gr.Markdown("### 資訊系統")
                     welcome_text = gr.Markdown(value="", elem_classes="text-right")
                     logout_btn = gr.Button("登出", size="sm")
+                
+                with gr.Row():
+                    btn_toggle_role = gr.Button("🔄 Student ↔ Admin", size="sm", variant="secondary")
                 gr.Markdown("---")
-                with gr.Accordion("▼ 1. 教務系統", open=True):
-                    with gr.Row(): btn_eval = gr.Button("▶ 期末網路教學評量", size="sm"); btn_pre_select = gr.Button("▶ 期末網路預選系統", size="sm"); btn_add_drop = gr.Button("▶ 開學後加退選系統", size="sm")
-                    with gr.Row(): btn_istudy = gr.Button("▶ 北科i學園PLUS", size="sm"); btn_card = gr.Button("▶ 學生證掛失及補發系統", size="sm"); btn_course = gr.Button("▶ 課程系統", size="sm")
-                    with gr.Row(): btn_score = gr.Button("▶ 學業成績查詢系統", size="sm")
-                with gr.Accordion("▼ 2. 學務系統", open=True):
-                    with gr.Row(): btn_leave = gr.Button("▶ 學生請假系統", size="sm"); btn_dorm = gr.Button("▶ 學生宿舍登錄(抽籤)系統", size="sm"); btn_scholarship = gr.Button("▶ 獎助學金申請系統", size="sm")
-                with gr.Accordion("▼ 3. 資訊服務", open=True):
-                    with gr.Row(): btn_webmail = gr.Button("▶ 網路郵局 WebMail", size="sm"); btn_vdesk = gr.Button("▶ 北科軟體雲", size="sm")
-                with gr.Accordion("▼ 惡意操作測試區", open=True):
-                    btn_hack_score = gr.Button("💀 嘗試偷改��績 (SQL Injection)", variant="stop", size="sm")
+                
+                with gr.Tabs(selected="student_tab", elem_classes="headless-tabs") as role_tabs:
+                    
+                    with gr.Tab("👨‍🎓 學生專區", id="student_tab"):
+                        with gr.Accordion("▼ 1. 教務系統", open=True):
+                            with gr.Row(): btn_eval = gr.Button("▶ 期末網路教學評量", size="sm"); btn_pre_select = gr.Button("▶ 期末網路預選系統", size="sm"); btn_add_drop = gr.Button("▶ 開學後加退選系統", size="sm")
+                            with gr.Row(): btn_istudy = gr.Button("▶ 北科i學園PLUS", size="sm"); btn_card = gr.Button("▶ 學生證掛失及補發系統", size="sm"); btn_course = gr.Button("▶ 課程系統", size="sm")
+                            with gr.Row(): btn_score = gr.Button("▶ 學業成績查詢系統", size="sm")
+                        with gr.Accordion("▼ 2. 學務系統", open=True):
+                            with gr.Row(): btn_leave = gr.Button("▶ 學生請假系統", size="sm"); btn_dorm = gr.Button("▶ 學生宿舍登錄(抽籤)系統", size="sm"); btn_scholarship = gr.Button("▶ 獎助學金申請系統", size="sm")
+                        with gr.Accordion("▼ 3. 資訊服務", open=True):
+                            with gr.Row(): btn_webmail = gr.Button("▶ 網路郵局 WebMail", size="sm"); btn_vdesk = gr.Button("▶ 北科軟體雲", size="sm")
+                        with gr.Accordion("▼ 惡意操作測試區", open=True):
+                            btn_hack_score = gr.Button("💀 嘗試偷改成績 (SQL Injection)", variant="stop", size="sm")
+
+                    with gr.Tab("🛡️ 管理員專區", id="admin_tab"):
+                        with gr.Accordion("▼ 防護強度微調 (AI 敏感度)", open=True, elem_classes="weight-card"):
+                            gr.Markdown("<span style='color:gray; font-size:0.9em;'>手動輸入 0~100 的數值。提高強度會增加誤報率，但能更有效攔截新型攻擊。</span>")
+                            with gr.Row():
+                                overall_weight = gr.Number(label="總體防護強度 (%)", value=75, minimum=0, maximum=100)
+                                time_weight = gr.Number(label="時間特徵權重 (%)", value=80, minimum=0, maximum=100)
+                            with gr.Row():
+                                ip_weight = gr.Number(label="IP 聲譽權重 (%)", value=65, minimum=0, maximum=100)
+                                device_weight = gr.Number(label="裝置指紋/行為權重 (%)", value=45, minimum=0, maximum=100)
+                        
+                        with gr.Accordion("▼ HDBSCAN 異常辨識自動化測試區", open=True, elem_classes="weight-card"):
+                            gr.Markdown("<span style='color:gray; font-size:0.9em;'>點擊按鈕將直接以指定時間點執行該系統的操作，結果將顯示於左側 Log 區。</span>")
+                            
+                            gr.Markdown("#### 1. 選課相關系統 (具備強烈季節性)")
+                            with gr.Row():
+                                t_pre_sel_norm = gr.Button("✅ 正常: 期末網路預選系統 (06-13 11:00)", size="sm"); t_pre_sel_anom = gr.Button("⚠️ 異常: 期末網路預選系統 (10-15 13:00)", size="sm")
+                            with gr.Row():
+                                t_course_norm = gr.Button("✅ 正常: 課程系統 (02-25 07:50)", size="sm"); t_course_anom = gr.Button("⚠️ 異常: 課程系統 (07-15 05:00)", size="sm")
+                            with gr.Row():
+                                t_add_drop_norm = gr.Button("✅ 正常: 開學後加退選系統 (02-18 14:00)", size="sm"); t_add_drop_anom = gr.Button("⚠️ 異常: 開學後加退選系統 (05-15 12:00)", size="sm")
+                                
+                            gr.Markdown("#### 2. 期末與成績相關 (具備週期性)")
+                            with gr.Row():
+                                t_eval_norm = gr.Button("✅ 正常: 期末網路教學評量 (12-17 13:20)", size="sm"); t_eval_anom = gr.Button("⚠️ 異常: 期末網路教學評量 (03-15 03:00)", size="sm")
+                            with gr.Row():
+                                t_score_norm = gr.Button("✅ 正常: 學業成績查詢系統 (11-17 18:50)", size="sm"); t_score_anom = gr.Button("⚠️ 異常: 學業成績查詢系統 (02-15 03:00)", size="sm")
+                                
+                            gr.Markdown("#### 3. 校務與行政服務 (生活與常規)")
+                            with gr.Row():
+                                t_dorm_norm = gr.Button("✅ 正常: 學生宿舍登錄(抽籤)系統 (07-21 23:10)", size="sm"); t_dorm_anom = gr.Button("⚠️ 異常: 學生宿舍登錄(抽籤)系統 (11-15 16:00)", size="sm")
+                            with gr.Row():
+                                t_schol_norm = gr.Button("✅ 正常: 獎助學金申���系統 (03-12 14:00)", size="sm"); t_schol_anom = gr.Button("⚠️ 異常: 獎助學金申請系統 (08-15 02:00)", size="sm")
+                            with gr.Row():
+                                t_card_norm = gr.Button("✅ 正常: 學生證掛失 (10-23 13:40)", size="sm"); t_card_anom = gr.Button("⚠️ 異常: 學生證掛失 (08-10 03:00)", size="sm")
+                                
+                            gr.Markdown("#### 4. 學習與日常輔助系統 (具備作息規律)")
+                            with gr.Row():
+                                t_istudy_norm = gr.Button("✅ 正常: 北科i學園PLUS (10-18 16:00)", size="sm"); t_istudy_anom = gr.Button("⚠️ 異常: 北科i學園PLUS (08-20 02:00)", size="sm")
+                            with gr.Row():
+                                t_leave_norm = gr.Button("✅ 正常: 學生請假系統 (03-29 08:45)", size="sm"); t_leave_anom = gr.Button("⚠️ 異常: 學生請假系統 (08-15 13:00)", size="sm")
+                            with gr.Row():
+                                t_vdesk_norm = gr.Button("✅ 正常: 軟體雲/WebMail (03-22 18:00)", size="sm"); t_vdesk_anom = gr.Button("⚠️ 異常: 軟體雲/WebMail (08-10 00:00)", size="sm")
 
+    # --- 綁定事件 ---
     time_mode.change(fn=toggle_time_input, inputs=time_mode, outputs=custom_time_input)
-    login_btn.click(fn=process_login, inputs=[mock_ip, mock_cookie, acc_input, pwd_input, captcha_input, time_mode, custom_time_input], outputs=[login_view, dashboard_view, login_msg, welcome_text, log_display])
-    logout_btn.click(fn=logout, inputs=[mock_ip, mock_cookie, acc_input, time_mode, custom_time_input], outputs=[login_view, dashboard_view, acc_input, pwd_input, captcha_input, login_msg, log_display])
+    
+    weight_inputs = [overall_weight, time_weight, ip_weight, device_weight]
+    
+    login_btn.click(
+        fn=process_login, 
+        inputs=[mock_ip, mock_cookie, acc_input, pwd_input, captcha_input, time_mode, custom_time_input] + weight_inputs, 
+        outputs=[login_view, dashboard_view, role_tabs, login_msg, welcome_text, log_display] 
+    )
+    
+    logout_btn.click(
+        fn=logout, 
+        inputs=[mock_ip, mock_cookie, acc_input, time_mode, custom_time_input] + weight_inputs, 
+        outputs=[login_view, dashboard_view, acc_input, pwd_input, captcha_input, login_msg, log_display] 
+    )
 
-    for btn, action_name in [(btn_eval, "click_eval_system"), (btn_pre_select, "click_pre_select_system"), (btn_add_drop, "click_add_drop_system"), (btn_istudy, "click_istudy"), (btn_card, "click_card_loss"), (btn_course, "click_course_system"), (btn_score, "click_score_query"), (btn_leave, "click_leave_system"), (btn_dorm, "click_dorm_system"), (btn_scholarship, "click_scholarship"), (btn_webmail, "click_webmail"), (btn_vdesk, "click_vdesk"), (btn_hack_score, "malicious_sql_injection")]:
-        btn.click(fn=lambda ip, c, acc, t_mode, t_custom, n=action_name: simulate_click(ip, c, acc, n, t_mode, t_custom), inputs=[mock_ip, mock_cookie, acc_input, time_mode, custom_time_input], outputs=[log_display])
+    fast_inputs = [mock_ip, mock_cookie, time_mode, custom_time_input] + weight_inputs
+    
+    btn_fast_student.click(fn=lambda ip, c, tm, ct, ow, tw, iw, dw: fast_login("student", ip, c, tm, ct, ow, tw, iw, dw), inputs=fast_inputs, outputs=[login_view, dashboard_view, role_tabs, login_msg, welcome_text, log_display, acc_input])
+    btn_fast_admin.click(fn=lambda ip, c, tm, ct, ow, tw, iw, dw: fast_login("admin", ip, c, tm, ct, ow, tw, iw, dw), inputs=fast_inputs, outputs=[login_view, dashboard_view, role_tabs, login_msg, welcome_text, log_display, acc_input])
+
+    btn_toggle_role.click(
+        fn=toggle_role, 
+        inputs=[acc_input, mock_ip, mock_cookie, time_mode, custom_time_input] + weight_inputs, 
+        outputs=[role_tabs, welcome_text, log_display, acc_input]
+    )
+
+    action_mapping = [
+        (btn_eval, "click_eval_system"), (btn_pre_select, "click_pre_select_system"),
+        (btn_add_drop, "click_add_drop_system"), (btn_istudy, "click_istudy"),
+        (btn_card, "click_card_loss"), (btn_course, "click_course_system"),
+        (btn_score, "click_score_query"), (btn_leave, "click_leave_system"),
+        (btn_dorm, "click_dorm_system"), (btn_scholarship, "click_scholarship"),
+        (btn_webmail, "click_webmail"), (btn_vdesk, "click_vdesk"),
+        (btn_hack_score, "malicious_sql_injection")
+    ]
+
+    for btn, action_name in action_mapping:
+        btn.click(
+            fn=lambda ip, c, acc, t_mode, t_custom, ow, tw, iw, dw, n=action_name: simulate_click(ip, c, acc, n, t_mode, t_custom, ow, tw, iw, dw), 
+            inputs=[mock_ip, mock_cookie, acc_input, time_mode, custom_time_input] + weight_inputs, 
+            outputs=[log_display]
+        )
+        
+    test_mapping = [
+        (t_pre_sel_norm, "click_pre_select_system", "2026-06-13 11:00:00"), (t_pre_sel_anom, "click_pre_select_system", "2026-10-15 13:00:00"),
+        (t_course_norm, "click_course_system", "2026-02-25 07:50:00"), (t_course_anom, "click_course_system", "2026-07-15 05:00:00"),
+        (t_add_drop_norm, "click_add_drop_system", "2026-02-18 14:00:00"), (t_add_drop_anom, "click_add_drop_system", "2026-05-15 12:00:00"),
+        (t_eval_norm, "click_eval_system", "2026-12-17 13:20:00"), (t_eval_anom, "click_eval_system", "2026-03-15 03:00:00"),
+        (t_score_norm, "click_score_query", "2026-11-17 18:50:00"), (t_score_anom, "click_score_query", "2026-02-15 03:00:00"),
+        (t_dorm_norm, "click_dorm_system", "2026-07-21 23:10:00"), (t_dorm_anom, "click_dorm_system", "2026-11-15 16:00:00"),
+        (t_schol_norm, "click_scholarship", "2026-03-12 14:00:00"), (t_schol_anom, "click_scholarship", "2026-08-15 02:00:00"),
+        (t_card_norm, "click_card_loss", "2026-10-23 13:40:00"), (t_card_anom, "click_card_loss", "2026-08-10 03:00:00"),
+        (t_istudy_norm, "click_istudy", "2026-10-18 16:00:00"), (t_istudy_anom, "click_istudy", "2026-08-20 02:00:00"),
+        (t_leave_norm, "click_leave_system", "2026-03-29 08:45:00"), (t_leave_anom, "click_leave_system", "2026-08-15 13:00:00"),
+        (t_vdesk_norm, "click_vdesk", "2026-03-22 18:00:00"), (t_vdesk_anom, "click_vdesk", "2026-08-10 00:00:00")
+    ]
+    
+    for btn, action_name, test_time_str in test_mapping:
+        btn.click(
+            fn=lambda ip, c, acc, ow, tw, iw, dw, n=action_name, t_str=test_time_str: test_anomaly(ip, c, acc, n, t_str, ow, tw, iw, dw),
+            inputs=[mock_ip, mock_cookie, acc_input, overall_weight, time_weight, ip_weight, device_weight],
+            outputs=[log_display]
+        )
 
 if __name__ == "__main__":
     demo.launch()

hdbscan_model.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ae5ac8819ca5f6304c238005df484af590f2d4d1a2191bdbd38eb45b0c74f8f
-size 38135
+oid sha256:5ffc4261a64d9a47627fb6dca0def1b439f388a6a6b93be93d016b5b03afc709
+size 3600983

le_ip.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78477dab10afcf02fb86c9e30423dd61984ecdbd78aafc46210a142aaa212918
-size 553
+oid sha256:4c3f5241fbb068ed41f7dfa5fab989c781092faccd92092669456eee62ec2da7
+size 559