|
|
@@ -53,6 +53,65 @@ def load_persona_library(json_path: str) -> Dict[str, Any]:
|
|
|
return json.load(f)
|
|
|
|
|
|
|
|
|
+def extract_all_features_from_how(how_json_path: str) -> Dict[str, Dict[str, Any]]:
|
|
|
+ """
|
|
|
+ 从how解构结果中提取所有特征的完整信息
|
|
|
+
|
|
|
+ Returns:
|
|
|
+ Dict[特征名称, {
|
|
|
+ 'similarity': float, # 最高相似度
|
|
|
+ 'weight': float, # 权重
|
|
|
+ 'dimension': str, # 所属维度
|
|
|
+ 'category': str # 分类：'已搜索', '待搜索', '低相似度'
|
|
|
+ }]
|
|
|
+ """
|
|
|
+ with open(how_json_path, 'r', encoding='utf-8') as f:
|
|
|
+ how_data = json.load(f)
|
|
|
+
|
|
|
+ features = {}
|
|
|
+
|
|
|
+ for dimension_key in ['灵感点列表', '目的点列表', '关键点列表']:
|
|
|
+ dimension_list = how_data.get('how解构结果', {}).get(dimension_key, [])
|
|
|
+ dimension_name = dimension_key.replace('列表', '')
|
|
|
+
|
|
|
+ for point in dimension_list:
|
|
|
+ feature_list = point.get('特征列表', [])
|
|
|
+
|
|
|
+ for feature in feature_list:
|
|
|
+ feature_name = feature.get('特征名称', '')
|
|
|
+ weight = feature.get('权重', 0)
|
|
|
+
|
|
|
+ # 查找最高相似度
|
|
|
+ max_similarity = 0
|
|
|
+ how_steps = point.get('how步骤列表', [])
|
|
|
+
|
|
|
+ for step in how_steps:
|
|
|
+ step_features = step.get('特征列表', [])
|
|
|
+ for step_feature in step_features:
|
|
|
+ if step_feature.get('特征名称') == feature_name:
|
|
|
+ matches = step_feature.get('匹配结果', [])
|
|
|
+ for match in matches:
|
|
|
+ similarity = match.get('匹配结果', {}).get('相似度', 0)
|
|
|
+ max_similarity = max(max_similarity, similarity)
|
|
|
+
|
|
|
+ # 确定分类
|
|
|
+ if max_similarity < 0.4:
|
|
|
+ category = '低相似度'
|
|
|
+ elif 0.4 <= max_similarity <= 0.8:
|
|
|
+ category = '待搜索' # 后续会根据Stage6数据更新为'已搜索'
|
|
|
+ else:
|
|
|
+ category = '高相似度' # >0.8，孤立点
|
|
|
+
|
|
|
+ features[feature_name] = {
|
|
|
+ 'similarity': max_similarity,
|
|
|
+ 'weight': weight,
|
|
|
+ 'dimension': dimension_name,
|
|
|
+ 'category': category
|
|
|
+ }
|
|
|
+
|
|
|
+ return features
|
|
|
+
|
|
|
+
|
|
|
def is_category_or_feature(persona_name: str, persona_data: Dict[str, Any]) -> str:
|
|
|
"""
|
|
|
递归判断人设项是特征还是分类
|
|
|
@@ -100,12 +159,16 @@ def is_category_or_feature(persona_name: str, persona_data: Dict[str, Any]) -> s
|
|
|
return 'feature'
|
|
|
|
|
|
|
|
|
-def extract_relationship_data(how_json_path: str, persona_data: Dict[str, Any]) -> List[Dict[str, Any]]:
|
|
|
+def extract_relationship_data(how_json_path: str, persona_data: Dict[str, Any]) -> tuple:
|
|
|
"""
|
|
|
从how解构结果中提取关系数据
|
|
|
|
|
|
Returns:
|
|
|
- List of relationship data with format:
|
|
|
+ Tuple of (relationships, all_high_matches):
|
|
|
+ - relationships: 最高相似度匹配的关系列表
|
|
|
+ - all_high_matches: 所有相似度>0.8的匹配列表
|
|
|
+
|
|
|
+ Data format:
|
|
|
{
|
|
|
'post_feature': str, # 帖子特征名称
|
|
|
'dimension': str, # 所属维度（灵感点/目的点/关键点）
|
|
|
@@ -119,6 +182,7 @@ def extract_relationship_data(how_json_path: str, persona_data: Dict[str, Any])
|
|
|
how_data = json.load(f)
|
|
|
|
|
|
relationships = []
|
|
|
+ all_high_matches = [] # 所有相似度>0.8的匹配
|
|
|
|
|
|
# 遍历how解构结果
|
|
|
for dimension_key in ['灵感点列表', '目的点列表', '关键点列表']:
|
|
|
@@ -144,6 +208,21 @@ def extract_relationship_data(how_json_path: str, persona_data: Dict[str, Any])
|
|
|
matches = step_feature.get('匹配结果', [])
|
|
|
for match in matches:
|
|
|
similarity = match.get('匹配结果', {}).get('相似度', 0)
|
|
|
+ persona_name = match.get('人设特征名称', '')
|
|
|
+
|
|
|
+ # 收集所有相似度>0.8的匹配
|
|
|
+ if similarity > 0.8:
|
|
|
+ item_type = is_category_or_feature(persona_name, persona_data)
|
|
|
+ all_high_matches.append({
|
|
|
+ 'post_feature': feature_name,
|
|
|
+ 'dimension': dimension_name,
|
|
|
+ 'weight': weight,
|
|
|
+ 'persona_item': persona_name,
|
|
|
+ 'similarity': similarity,
|
|
|
+ 'item_type': item_type
|
|
|
+ })
|
|
|
+
|
|
|
+ # 追踪最高相似度
|
|
|
if similarity > max_similarity:
|
|
|
max_similarity = similarity
|
|
|
max_match = match
|
|
|
@@ -162,15 +241,17 @@ def extract_relationship_data(how_json_path: str, persona_data: Dict[str, Any])
|
|
|
'item_type': item_type
|
|
|
})
|
|
|
|
|
|
- return relationships
|
|
|
+ return relationships, all_high_matches
|
|
|
|
|
|
|
|
|
-def generate_relationship_graph_html(relationships: List[Dict[str, Any]]) -> str:
|
|
|
+def generate_relationship_graph_html(relationships: List[Dict[str, Any]],
|
|
|
+ all_high_matches: List[Dict[str, Any]]) -> str:
|
|
|
"""
|
|
|
生成关系图的SVG HTML代码
|
|
|
|
|
|
Args:
|
|
|
- relationships: extract_relationship_data返回的关系数据列表
|
|
|
+ relationships: 最高相似度匹配的关系数据列表
|
|
|
+ all_high_matches: 所有相似度>0.8的匹配列表
|
|
|
|
|
|
Returns:
|
|
|
HTML字符串（包含SVG）
|
|
|
@@ -178,8 +259,18 @@ def generate_relationship_graph_html(relationships: List[Dict[str, Any]]) -> str
|
|
|
if not relationships:
|
|
|
return '<div style="padding: 40px; text-align: center; color: #6b7280;">暂无关系数据</div>'
|
|
|
|
|
|
- # 提取唯一的人设项和帖子特征
|
|
|
- persona_items = list({(r['persona_item'], r['item_type'], r['similarity']) for r in relationships})
|
|
|
+ # 合并所有需要显示的人设项（去重）
|
|
|
+ all_personas = set()
|
|
|
+
|
|
|
+ # 添加relationships中的人设项
|
|
|
+ for r in relationships:
|
|
|
+ all_personas.add((r['persona_item'], r['item_type'], r['similarity']))
|
|
|
+
|
|
|
+ # 添加所有>0.8的人设项
|
|
|
+ for match in all_high_matches:
|
|
|
+ all_personas.add((match['persona_item'], match['item_type'], match['similarity']))
|
|
|
+
|
|
|
+ persona_items = list(all_personas)
|
|
|
|
|
|
# 为每个帖子特征关联最高相似度
|
|
|
post_feature_map = {}
|
|
|
@@ -266,13 +357,44 @@ def generate_relationship_graph_html(relationships: List[Dict[str, Any]]) -> str
|
|
|
</marker>
|
|
|
</defs>
|
|
|
|
|
|
+ <!-- 左右标记 -->
|
|
|
+ <g class="labels">
|
|
|
+ <!-- 人设标记 -->
|
|
|
+ <text x="{left_margin}" y="30"
|
|
|
+ font-size="18" font-weight="600" fill="#6b7280"
|
|
|
+ text-anchor="middle">人设</text>
|
|
|
+
|
|
|
+ <!-- 帖子标记 -->
|
|
|
+ <text x="{left_margin + middle_space}" y="30"
|
|
|
+ font-size="18" font-weight="600" fill="#6b7280"
|
|
|
+ text-anchor="middle">帖子</text>
|
|
|
+ </g>
|
|
|
+
|
|
|
<!-- 连接线 -->
|
|
|
<g class="connections">
|
|
|
''']
|
|
|
|
|
|
+ # 合并所有需要绘制的连接（去重）
|
|
|
+ all_connections = []
|
|
|
+ connection_keys = set()
|
|
|
+
|
|
|
+ # 添加主要关系（最高相似度匹配）
|
|
|
+ for rel in relationships:
|
|
|
+ key = (rel['persona_item'], rel['post_feature'])
|
|
|
+ if key not in connection_keys:
|
|
|
+ all_connections.append(rel)
|
|
|
+ connection_keys.add(key)
|
|
|
+
|
|
|
+ # 添加额外的>0.8关系
|
|
|
+ for match in all_high_matches:
|
|
|
+ key = (match['persona_item'], match['post_feature'])
|
|
|
+ if key not in connection_keys:
|
|
|
+ all_connections.append(match)
|
|
|
+ connection_keys.add(key)
|
|
|
+
|
|
|
# 绘制连接线
|
|
|
line_idx = 0 # 计数器用于标签错开
|
|
|
- for rel in relationships:
|
|
|
+ for rel in all_connections:
|
|
|
persona_pos = persona_positions.get(rel['persona_item'])
|
|
|
post_pos = post_positions.get(rel['post_feature'])
|
|
|
|
|
|
@@ -484,6 +606,8 @@ def calculate_statistics(data: List[Dict[str, Any]]) -> Dict[str, Any]:
|
|
|
def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
stage7_mapping: Dict[str, Any], stage8_mapping: Dict[str, Any],
|
|
|
relationship_graph_html: str,
|
|
|
+ all_features: Dict[str, Dict[str, Any]],
|
|
|
+ low_similarity_features: List[Dict[str, Any]],
|
|
|
output_path: str):
|
|
|
"""生成HTML可视化页面"""
|
|
|
|
|
|
@@ -491,6 +615,8 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
data_json = json.dumps(data, ensure_ascii=False, indent=2)
|
|
|
stage7_json = json.dumps(stage7_mapping, ensure_ascii=False, indent=2)
|
|
|
stage8_json = json.dumps(stage8_mapping, ensure_ascii=False, indent=2)
|
|
|
+ all_features_json = json.dumps(all_features, ensure_ascii=False, indent=2)
|
|
|
+ low_similarity_json = json.dumps(low_similarity_features, ensure_ascii=False, indent=2)
|
|
|
|
|
|
html_content = f'''<!DOCTYPE html>
|
|
|
<html lang="zh-CN">
|
|
|
@@ -2159,6 +2285,117 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
.connection-line.dimmed {{
|
|
|
opacity: 0.1 !important;
|
|
|
}}
|
|
|
+
|
|
|
+ /* ========== 低相似度特征样式 ========== */
|
|
|
+ .low-similarity-section {{
|
|
|
+ margin-top: 20px;
|
|
|
+ border-top: 2px solid #fca5a5;
|
|
|
+ padding-top: 15px;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-similarity-header {{
|
|
|
+ display: flex;
|
|
|
+ align-items: center;
|
|
|
+ justify-content: space-between;
|
|
|
+ padding: 12px 15px;
|
|
|
+ background: linear-gradient(135deg, #fee2e2 0%, #fef2f2 100%);
|
|
|
+ border-radius: 8px;
|
|
|
+ cursor: pointer;
|
|
|
+ border: 2px solid #fca5a5;
|
|
|
+ transition: all 0.3s ease;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-similarity-header:hover {{
|
|
|
+ background: linear-gradient(135deg, #fecaca 0%, #fee2e2 100%);
|
|
|
+ border-color: #f87171;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-similarity-title {{
|
|
|
+ font-size: 14px;
|
|
|
+ font-weight: 600;
|
|
|
+ color: #dc2626;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-similarity-count {{
|
|
|
+ font-size: 12px;
|
|
|
+ font-weight: 600;
|
|
|
+ color: #991b1b;
|
|
|
+ background: white;
|
|
|
+ padding: 2px 8px;
|
|
|
+ border-radius: 12px;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .toggle-icon {{
|
|
|
+ font-size: 12px;
|
|
|
+ color: #dc2626;
|
|
|
+ transition: transform 0.3s ease;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .toggle-icon.rotated {{
|
|
|
+ transform: rotate(180deg);
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-similarity-list {{
|
|
|
+ margin-top: 10px;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-similarity-item {{
|
|
|
+ padding: 10px 12px;
|
|
|
+ margin: 8px 0;
|
|
|
+ background: white;
|
|
|
+ border: 2px solid #fecaca;
|
|
|
+ border-radius: 6px;
|
|
|
+ cursor: pointer;
|
|
|
+ transition: all 0.2s ease;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-similarity-item:hover {{
|
|
|
+ border-color: #f87171;
|
|
|
+ background: #fef2f2;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-feature-header {{
|
|
|
+ display: flex;
|
|
|
+ align-items: center;
|
|
|
+ justify-content: space-between;
|
|
|
+ margin-bottom: 4px;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-feature-name {{
|
|
|
+ font-size: 13px;
|
|
|
+ font-weight: 600;
|
|
|
+ color: #991b1b;
|
|
|
+ flex: 1;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-feature-score {{
|
|
|
+ font-size: 14px;
|
|
|
+ font-weight: 700;
|
|
|
+ color: #dc2626;
|
|
|
+ background: #fee2e2;
|
|
|
+ padding: 2px 8px;
|
|
|
+ border-radius: 4px;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-feature-meta {{
|
|
|
+ font-size: 11px;
|
|
|
+ color: #9ca3af;
|
|
|
+ margin-bottom: 6px;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-feature-detail {{
|
|
|
+ padding: 8px;
|
|
|
+ background: #fef2f2;
|
|
|
+ border-radius: 4px;
|
|
|
+ border-left: 3px solid #f87171;
|
|
|
+ margin-top: 8px;
|
|
|
+ }}
|
|
|
+
|
|
|
+ .low-feature-reason {{
|
|
|
+ font-size: 12px;
|
|
|
+ color: #7f1d1d;
|
|
|
+ line-height: 1.6;
|
|
|
+ }}
|
|
|
</style>
|
|
|
</head>
|
|
|
<body>
|
|
|
@@ -2241,7 +2478,7 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
<!-- Tab导航 -->
|
|
|
<div class="tab-navigation">
|
|
|
<button class="tab-button active" onclick="switchTab('search')">📋 搜索结果</button>
|
|
|
- <button class="tab-button" onclick="switchTab('relationship')">📊 关系图</button>
|
|
|
+ <button class="tab-button" onclick="switchTab('relationship')">📊 人设-帖子匹配</button>
|
|
|
</div>
|
|
|
|
|
|
<!-- Tab内容区域 -->
|
|
|
@@ -2261,7 +2498,7 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
<!-- Tab2: 关系图 -->
|
|
|
<div class="tab-pane" id="tab-relationship">
|
|
|
<div class="relationship-section">
|
|
|
- <div class="section-header">📊 帖子-人设关系图</div>
|
|
|
+ <div class="section-header">📊 人设-帖子关系图</div>
|
|
|
{relationship_graph_html}
|
|
|
</div>
|
|
|
</div>
|
|
|
@@ -2310,6 +2547,8 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
const data = {data_json};
|
|
|
const stage7Data = {stage7_json};
|
|
|
const stage8Data = {stage8_json};
|
|
|
+ const allFeatures = {all_features_json};
|
|
|
+ const lowSimilarityFeatures = {low_similarity_json};
|
|
|
let currentFilter = 'all';
|
|
|
|
|
|
// Tab切换功能
|
|
|
@@ -2431,10 +2670,20 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
totalSearches += (group['top10_searches'] || []).length;
|
|
|
}});
|
|
|
|
|
|
+ // 从allFeatures获取特征信息
|
|
|
+ const featureName = feature['原始特征名称'];
|
|
|
+ const featureInfo = allFeatures[featureName] || {{}};
|
|
|
+ const similarity = featureInfo.similarity || 0;
|
|
|
+ const weight = featureInfo.weight || 0;
|
|
|
+ const dimension = featureInfo.dimension || '';
|
|
|
+
|
|
|
html += `
|
|
|
<div class="feature-group">
|
|
|
<div class="feature-header" onclick="toggleFeature(${{featureIdx}})" id="feature-header-${{featureIdx}}">
|
|
|
- <div class="feature-title post-target-word">📝 ${{feature['原始特征名称']}}</div>
|
|
|
+ <div class="feature-title post-target-word">📝 ${{featureName}}</div>
|
|
|
+ <div class="feature-meta" style="font-size:11px;color:#9ca3af;margin-top:4px;">
|
|
|
+ 相似度: ${{similarity.toFixed(2)}} · ${{dimension}}
|
|
|
+ </div>
|
|
|
</div>
|
|
|
<div class="search-words-list" id="search-words-${{featureIdx}}">
|
|
|
`;
|
|
|
@@ -2455,7 +2704,7 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
<div class="base-word-meta">相似度: ${{baseSimilarity.toFixed(2)}} · ${{searches.length}}个搜索词</div>
|
|
|
</div>
|
|
|
<div class="base-word-desc" id="base-word-desc-${{featureIdx}}-${{groupIdx}}">
|
|
|
- ${{relatedWordNames || '无相关词汇'}}
|
|
|
+ <span style="color:#92400e;font-weight:600;">来源人设候选特征/分类: </span>${{relatedWordNames || '无相关词汇'}}
|
|
|
</div>
|
|
|
<div class="search-words-sublist" id="search-words-sublist-${{featureIdx}}-${{groupIdx}}">
|
|
|
`;
|
|
|
@@ -2533,6 +2782,49 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
`;
|
|
|
}});
|
|
|
|
|
|
+ // 添加低相似度特征板块
|
|
|
+ if (lowSimilarityFeatures && lowSimilarityFeatures.length > 0) {{
|
|
|
+ html += `
|
|
|
+ <div class="low-similarity-section">
|
|
|
+ <div class="low-similarity-header" onclick="toggleLowSimilarity()">
|
|
|
+ <div class="low-similarity-title">🔴 低相似度特征（未搜索）</div>
|
|
|
+ <div class="low-similarity-count">${{lowSimilarityFeatures.length}}个</div>
|
|
|
+ <div class="toggle-icon" id="low-similarity-toggle">▼</div>
|
|
|
+ </div>
|
|
|
+ <div class="low-similarity-list" id="low-similarity-list" style="display:none;">
|
|
|
+ `;
|
|
|
+
|
|
|
+ lowSimilarityFeatures.forEach((feature, idx) => {{
|
|
|
+ const name = feature.name || '';
|
|
|
+ const similarity = feature.similarity || 0;
|
|
|
+ const weight = feature.weight || 0;
|
|
|
+ const dimension = feature.dimension || '';
|
|
|
+
|
|
|
+ html += `
|
|
|
+ <div class="low-similarity-item" onclick="toggleLowFeatureDetail(${{idx}})" id="low-feature-${{idx}}">
|
|
|
+ <div class="low-feature-header">
|
|
|
+ <div class="low-feature-name">📝 ${{name}}</div>
|
|
|
+ <div class="low-feature-score">${{similarity.toFixed(2)}}</div>
|
|
|
+ </div>
|
|
|
+ <div class="low-feature-meta">
|
|
|
+ ${{dimension}}
|
|
|
+ </div>
|
|
|
+ <div class="low-feature-detail" id="low-feature-detail-${{idx}}" style="display:none;">
|
|
|
+ <div class="low-feature-reason">
|
|
|
+ ❌ 相似度 ${{similarity.toFixed(2)}} < 0.4<br>
|
|
|
+ 低于搜索阈值，未执行搜索
|
|
|
+ </div>
|
|
|
+ </div>
|
|
|
+ </div>
|
|
|
+ `;
|
|
|
+ }});
|
|
|
+
|
|
|
+ html += `
|
|
|
+ </div>
|
|
|
+ </div>
|
|
|
+ `;
|
|
|
+ }}
|
|
|
+
|
|
|
sidebar.innerHTML = html;
|
|
|
}}
|
|
|
|
|
|
@@ -3194,6 +3486,35 @@ def generate_html(data: List[Dict[str, Any]], stats: Dict[str, Any],
|
|
|
}}
|
|
|
}}
|
|
|
|
|
|
+ // 切换低相似度特征板块展开/折叠
|
|
|
+ function toggleLowSimilarity() {{
|
|
|
+ const list = document.getElementById('low-similarity-list');
|
|
|
+ const toggle = document.getElementById('low-similarity-toggle');
|
|
|
+
|
|
|
+ if (list && toggle) {{
|
|
|
+ if (list.style.display === 'none') {{
|
|
|
+ list.style.display = 'block';
|
|
|
+ toggle.classList.add('rotated');
|
|
|
+ }} else {{
|
|
|
+ list.style.display = 'none';
|
|
|
+ toggle.classList.remove('rotated');
|
|
|
+ }}
|
|
|
+ }}
|
|
|
+ }}
|
|
|
+
|
|
|
+ // 切换单个低相似度特征的详情展开/折叠
|
|
|
+ function toggleLowFeatureDetail(idx) {{
|
|
|
+ const detail = document.getElementById(`low-feature-detail-${{idx}}`);
|
|
|
+
|
|
|
+ if (detail) {{
|
|
|
+ if (detail.style.display === 'none') {{
|
|
|
+ detail.style.display = 'block';
|
|
|
+ }} else {{
|
|
|
+ detail.style.display = 'none';
|
|
|
+ }}
|
|
|
+ }}
|
|
|
+ }}
|
|
|
+
|
|
|
function filterNotes(category) {{
|
|
|
currentFilter = category;
|
|
|
|
|
|
@@ -3459,7 +3780,7 @@ def main():
|
|
|
print(f" - 完全匹配: {stats['match_complete']} ({stats['complete_rate']}%)")
|
|
|
|
|
|
# 生成关系图
|
|
|
- print(f"\n📊 生成帖子-人设关系图...")
|
|
|
+ print(f"\n📊 生成人设-帖子关系图...")
|
|
|
relationship_graph_html = ""
|
|
|
try:
|
|
|
print(f"📖 加载人设库: {persona_library_path}")
|
|
|
@@ -3467,10 +3788,11 @@ def main():
|
|
|
print(f"✓ 加载人设库完成")
|
|
|
|
|
|
print(f"📖 提取关系数据: {how_json_path}")
|
|
|
- relationships = extract_relationship_data(how_json_path, persona_data)
|
|
|
- print(f"✓ 提取了 {len(relationships)} 个关系")
|
|
|
+ relationships, all_high_matches = extract_relationship_data(how_json_path, persona_data)
|
|
|
+ print(f"✓ 提取了 {len(relationships)} 个最高匹配关系")
|
|
|
+ print(f"✓ 提取了 {len(all_high_matches)} 个高相似度(>0.8)匹配")
|
|
|
|
|
|
- relationship_graph_html = generate_relationship_graph_html(relationships)
|
|
|
+ relationship_graph_html = generate_relationship_graph_html(relationships, all_high_matches)
|
|
|
print(f"✓ 关系图生成完成")
|
|
|
except FileNotFoundError as e:
|
|
|
print(f"⚠️ 警告: 无法生成关系图 - {e}")
|
|
|
@@ -3479,8 +3801,36 @@ def main():
|
|
|
print(f"⚠️ 警告: 生成关系图时出错 - {e}")
|
|
|
relationship_graph_html = f'<div style="padding: 40px; text-align: center; color: #ef4444;">关系图生成失败: {e}</div>'
|
|
|
|
|
|
+ # 提取所有特征信息（包括低相似度特征）
|
|
|
+ print(f"\n📊 提取所有特征信息...")
|
|
|
+ all_features = {}
|
|
|
+ low_similarity_features = []
|
|
|
+ try:
|
|
|
+ all_features = extract_all_features_from_how(how_json_path)
|
|
|
+
|
|
|
+ # 标记Stage6中已搜索的特征
|
|
|
+ stage6_feature_names = set([item.get('原始特征名称') for item in data])
|
|
|
+ for feature_name in stage6_feature_names:
|
|
|
+ if feature_name in all_features and all_features[feature_name]['category'] == '待搜索':
|
|
|
+ all_features[feature_name]['category'] = '已搜索'
|
|
|
+
|
|
|
+ # 提取低相似度特征（<0.4）
|
|
|
+ low_similarity_features = [
|
|
|
+ {'name': name, **info}
|
|
|
+ for name, info in all_features.items()
|
|
|
+ if info['category'] == '低相似度'
|
|
|
+ ]
|
|
|
+ # 按相似度降序排序
|
|
|
+ low_similarity_features.sort(key=lambda x: x['similarity'], reverse=True)
|
|
|
+
|
|
|
+ print(f"✓ 提取了 {len(all_features)} 个特征")
|
|
|
+ print(f" - 低相似度特征: {len(low_similarity_features)} 个")
|
|
|
+ except Exception as e:
|
|
|
+ print(f"⚠️ 警告: 提取特征信息时出错 - {e}")
|
|
|
+ low_similarity_features = []
|
|
|
+
|
|
|
print(f"\n🎨 生成可视化页面...")
|
|
|
- generate_html(data, stats, stage7_mapping, stage8_mapping, relationship_graph_html, output_path)
|
|
|
+ generate_html(data, stats, stage7_mapping, stage8_mapping, relationship_graph_html, all_features, low_similarity_features, output_path)
|
|
|
print(f"✓ 生成完成: {output_path}")
|
|
|
|
|
|
print(f"\n🌐 在浏览器中打开查看:")
|
刘立冬