实现工程名和项目名规范化和多轮

api/constants.py

@@ -0,0 +1,21 @@
+# constants.py
+#日期
+DATE = "date"
+#工程名称
+PROJECT_NAME = "projectName"
+#工程性质
+PROJECT_TYPE = "projectType"
+#建管单位
+CONSTRUCTION_UNIT = "constructionUnit"
+#分公司
+IMPLEMENTATION_ORG = "implementationOrganization"
+#项目部
+PROJECT_DEPARTMENT = "projectDepartment"
+#项目经理
+PROJECT_MANAGER = "projectManager"
+#分包商
+SUBCONTRACTOR = "subcontractor"
+#班组
+TEAM_LEADER = "teamLeader"
+#风险
+RISK_LEVEL = "riskLevel"

api/mian.py

@@ -12,7 +12,8 @@ from pydantic import ValidationError
 from intentRecognition import IntentRecognition
 from slotRecognition import SlotRecognition
 from fuzzywuzzy import process
-import utils
+from utils import CheckResult, StandardType
+from constants import  PROJECT_NAME,  PROJECT_DEPARTMENT
 
 # 常量
 MODEL_ERNIE_PATH = R"../ernie/output/checkpoint-4160"
@@ -199,28 +200,26 @@ def agent():
             entities = slot_recognizer.recognize(query)
 
             print(f"意图识别后的label:{predicted_label}, id:{predicted_id},槽位抽取后的实体：{entities},message:{messages}")
+
             #必须槽位缺失检查
             status, sk = check_lost(predicted_id, entities)
-            if status == 1:
+            if status == CheckResult.NEEDS_MORE_ROUNDS:
+                return jsonify({"code": 10001, "msg": "成功",
+                                "answer": { "miss": sk},
+                                })
+
+            #工程名和项目名标准化
+            result, information = check_project_standard_slot(predicted_id, entities)
+            if result == CheckResult.NEEDS_MORE_ROUNDS:
                 return jsonify({
-                    "code": 200,
+                    "code": 10001, "msg": "成功",
-                    "msg": "成功",
+                    "answer": {"miss": information},
-                    "answer": {
-                        "miss": sk
-                    },
                 })
 
-            else:
+            return jsonify({
-                return jsonify({
+                "code": 200,"msg": "成功",
-                    "code": 200,
+                "answer": {"int": predicted_id, "label": predicted_label, "probability": predicted_probability, "slot": entities },
-                    "msg": "成功",
+            })
-                    "answer": {
-                        "int": predicted_id,
-                        "label": predicted_label,
-                        "probability": predicted_probability,
-                        "slot": entities
-                    },
-                })
 
         # 如果是后续轮次（多轮对话），这里只做示例，可能需要根据具体需求进行处理
         else:
@@ -248,39 +247,64 @@ def check_lost(int_res, slot):
         7: [['date']],
         8: [['date']],
     }
-    #3:"页面切换",
+
     intention_mapping = {2: "页面切换", 3: "日计划数量查询", 4: "周计划数量查询", 5: "日计划作业内容",
                         6: "周计划作业内容",7: "施工人数",8: "作业考勤人数"}
     if not mapping.__contains__(int_res):
         return 0, ""
+    #提取的槽位信息
     cur_k = list(slot.keys())
     idx = -1
     idx_len = 99
     for i in range(len(mapping[int_res])):
         sk = mapping[int_res][i]
-        left = [x for x in sk if x not in cur_k]
+        #不在提取的槽位信息里，但是在必须槽位表里
-        more = [x for x in cur_k if x not in sk]
+        miss_params = [x for x in sk if x not in cur_k]
-        if len(more) >= 0 and len(left) == 0:
+        #不在必须槽位表里，但是在提取的槽位信息里
+        extra_params = [x for x in cur_k if x not in sk]
+        if len(extra_params) >= 0 and len(miss_params) == 0:
             idx = i
             idx_len = 0
             break
-        if len(left) < idx_len:
+        if len(miss_params) < idx_len:
             idx = i
-            idx_len = len(left)
+            idx_len = len(miss_params)
 
     if idx_len == 0:  # 匹配通过
-        return 0, cur_k
+        return CheckResult.NO_MATCH, cur_k
+    #符合当前意图的的必须槽位，但是不在提取的槽位信息里
     left = [x for x in mapping[int_res][idx] if x not in cur_k]
+    print(f"符合当前意图的的必须槽位，但是不在提取的槽位信息里, {left}")
     apologize_str = "非常抱歉，"
     if int_res == 2:
-        return 1, f"{apologize_str}请问你想查询哪个页面？"
+        return CheckResult.NEEDS_MORE_ROUNDS, f"{apologize_str}请问你想查询哪个页面？"
     elif int_res in [3, 4, 5, 6, 7, 8]:
-        return 1, f"{apologize_str}请问你想查询什么时间的{intention_mapping[int_res]}？"
+        return CheckResult.NEEDS_MORE_ROUNDS, f"{apologize_str}请问你想查询什么时间的{intention_mapping[int_res]}？"
 
 
 #标准化工程名
-def check_project_standard_slot(int_res, slot)->tuple:
+def check_project_standard_slot(int_res, slot) -> tuple:
-    return 0, ''
+    intention_list = {3, 4, 5, 6, 7, 8}
+    if int_res not in intention_list:
+        return CheckResult.NO_MATCH,""
+
+    for key, value in slot.items():
+        if key == PROJECT_NAME:
+            match_project, match_possibility = fuzzy_match(value, standard_project_name_list)
+            if match_possibility >= 0.9:
+                slot[key] = match_project
+            else:
+                return CheckResult.NEEDS_MORE_ROUNDS, f"抱歉，您说的工程名是{match_project}吗"
+
+        if key == PROJECT_DEPARTMENT:
+            match_program, match_possibility = fuzzy_match(value, standard_program_name_list)
+            if match_possibility >= 0.9:
+                slot[key] = match_program
+            else:
+                return CheckResult.NEEDS_MORE_ROUNDS, f"抱歉，您说的项目名是{match_program}吗"
+
+    return CheckResult.NO_MATCH,""
+
 
 def fuzzy_match(user_input, standard_name):
     result = process.extract(user_input, standard_name)

api/utils.py

@@ -1,3 +1,4 @@
+from enum import Enum
 def load_standard_name(file_path:str):
     try:
         # f = open(file_path, 'r', encoding='utf-8')
@@ -16,3 +17,16 @@ def load_standard_name(file_path:str):
     except Exception as e:
         print(f"读取文件时发生错误：{e}")
         raise Exception(f"错误：文件 {file_path} 不存在")
+
+
+class CheckResult(Enum):
+    NO_MATCH = 0    # 不符合检查条件
+    MATCH_FOUND = 1  # 匹配到了值
+    NEEDS_MORE_ROUNDS = 2  # 需要多轮
+
+
+class StandardType(Enum):
+    #工程名检查
+    PROJECT_CHECK = 0
+    #项目名检查
+    PROGRAM_CHECK = 1