Update smart_breed_matcher.py

smart_breed_matcher.py

@@ -245,37 +245,86 @@ class SmartBreedMatcher:
         
         return similarity
 
+    # def _general_matching(self, description: str, top_n: int = 10) -> List[Dict]:
+    #     """基本的品種匹配邏輯"""
+    #     matches = []
+    #     for breed in self.dog_data:
+    #         breed_name = breed[1]
+    #         breed_description = breed[9]
+    #         temperament = breed[4]
+            
+    #         # 計算相似度
+    #         desc_embedding = self.model.encode(description)
+    #         breed_desc_embedding = self.model.encode(breed_description)
+    #         breed_temp_embedding = self.model.encode(temperament)
+            
+    #         # 計算描述和性格的相似度
+    #         desc_similarity = float(util.pytorch_cos_sim(desc_embedding, breed_desc_embedding))
+    #         temp_similarity = float(util.pytorch_cos_sim(desc_embedding, breed_temp_embedding))
+            
+    #         # 結合分數
+    #         final_score = (desc_similarity * 0.6 + temp_similarity * 0.4)
+            
+    #         matches.append({
+    #             'breed': breed_name,
+    #             'score': final_score,
+    #             'is_preferred': False,
+    #             'similarity': final_score,
+    #             'reason': "Matched based on general description and temperament"
+    #         })
+        
+    #     # 排序並返回前 N 個匹配結果
+    #     return sorted(matches, key=lambda x: -x['score'])[:top_n]
+
     def _general_matching(self, description: str, top_n: int = 10) -> List[Dict]:
-        """基本的品種匹配邏輯"""
+        """基本的品種匹配邏輯，考慮描述、性格、噪音和健康因素"""
         matches = []
         for breed in self.dog_data:
             breed_name = breed[1]
             breed_description = breed[9]
             temperament = breed[4]
             
-            # 計算相似度
+            # 計算描述文本和性格的相似度
             desc_embedding = self.model.encode(description)
             breed_desc_embedding = self.model.encode(breed_description)
             breed_temp_embedding = self.model.encode(temperament)
             
-            # 計算描述和性格的相似度
             desc_similarity = float(util.pytorch_cos_sim(desc_embedding, breed_desc_embedding))
             temp_similarity = float(util.pytorch_cos_sim(desc_embedding, breed_temp_embedding))
             
-            # 結合分數
-            final_score = (desc_similarity * 0.6 + temp_similarity * 0.4)
+            # 計算噪音相似度和健康分數
+            noise_similarity = self._calculate_noise_similarity(breed_name, breed_name)
+            health_score = self._calculate_health_score(breed_name)
+            health_similarity = 1.0 - abs(health_score - 0.8)  # 假設理想健康分數為 0.8
+            
+            # 加權計算分數
+            weights = {
+                'description': 0.35,
+                'temperament': 0.25,
+                'noise': 0.2,
+                'health': 0.2
+            }
+    
+            # 計算最終分數
+            final_score = (
+                desc_similarity * weights['description'] +
+                temp_similarity * weights['temperament'] +
+                noise_similarity * weights['noise'] +
+                health_similarity * weights['health']
+            )
             
             matches.append({
                 'breed': breed_name,
                 'score': final_score,
                 'is_preferred': False,
                 'similarity': final_score,
-                'reason': "Matched based on general description and temperament"
+                'reason': "Matched based on description, temperament, noise level, and health score"
             })
         
         # 排序並返回前 N 個匹配結果
         return sorted(matches, key=lambda x: -x['score'])[:top_n]
 
+
     def match_user_preference(self, description: str, top_n: int = 10) -> List[Dict]:
         """根據用戶描述匹配最適合的品種"""
         preferred_breed = self._detect_breed_preference(description)