Menu
处理社会网络数据，包括关系数据收集、矩阵构建、数据清洗验证和多模网络处理。当需要从问卷、访谈、观察或数字记录中提取关系数据，构建标准化的网络矩阵时使用此技能。
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Based on SOC occupation classification
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/ptreezh/sscisubagent-skills --skill processing-network-dataThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Showing SKILL.md
| name | processing-network-data |
| description | 处理社会网络数据，包括关系数据收集、矩阵构建、数据清洗验证和多模网络处理。当需要从问卷、访谈、观察或数字记录中提取关系数据，构建标准化的网络矩阵时使用此技能。 |
专门用于社会网络研究的数据收集和预处理，将原始关系数据转换为标准化的网络分析格式。
当用户提到以下需求时，使用此技能：
识别数据来源类型
数据格式标准化
数据导入和初步检查
# 示例：不同格式数据的导入
if data_format == 'excel':
df = pd.read_excel(file_path)
elif data_format == 'csv':
df = pd.read_csv(file_path, encoding='utf-8')
elif data_format == 'json':
with open(file_path, 'r', encoding='utf-8') as f:
data = json.load(f)
问卷关系提取
访谈文本关系挖掘
# 从访谈文本中提取关系
def extract_relationships_from_text(text):
# 使用正则表达式和关键词匹配
relationship_patterns = [
r'(\w+)\s*[和|与]\s*(\w+)\s*[是|为|有].*关系',
r'(\w+)\s*[认识|了解]\s*(\w+)',
r'(\w+)\s*[向|对]\s*(\w+)\s*[寻求|请求].*帮助'
]
relationships = []
for pattern in relationship_patterns:
matches = re.findall(pattern, text)
relationships.extend(matches)
return relationships
观察数据关系编码
数字数据关系挖掘
确定节点列表
# 构建完整的节点列表
def create_node_list(relationship_data):
nodes = set()
for rel in relationship_data:
nodes.add(rel['source'])
nodes.add(rel['target'])
return sorted(list(nodes))
构建邻接矩阵
# 构建邻接矩阵
def build_adjacency_matrix(nodes, relationships, weighted=False):
n = len(nodes)
adjacency_matrix = np.zeros((n, n))
node_index = {node: i for i, node in enumerate(nodes)}
for rel in relationships:
i = node_index[rel['source']]
j = node_index[rel['target']]
if weighted:
adjacency_matrix[i][j] = rel.get('weight', 1)
else:
adjacency_matrix[i][j] = 1
if not rel.get('directed', False):
adjacency_matrix[j][i] = 1
return adjacency_matrix
处理特殊网络类型
缺失值处理
异常值检测
# 异常关系检测
def detect_outlier_relationships(adjacency_matrix, threshold=3):
degrees = np.sum(adjacency_matrix, axis=1)
mean_degree = np.mean(degrees)
std_degree = np.std(degrees)
outliers = []
for i, degree in enumerate(degrees):
if abs(degree - mean_degree) > threshold * std_degree:
outliers.append(i)
return outliers
数据一致性检查
质量评估指标
节点属性整合
边属性处理
时间属性编码
问题：问卷数据格式不统一
问题：访谈文本关系识别困难
问题：数据质量参差不齐
问题：多源数据整合困难
高质量的网络数据处理应该：
此技能为中文社会科学网络研究提供全面的数据处理支持，确保从原始数据到标准化网络文件的准确转换。