越览（128）——精读期刊论文的2方法原理（1）

摘要：This issue will introduce the principle of the method of the intensively read replica paper "Crowd intelligence knowledge mining m

分享兴趣，传播快乐，

增长见闻，留下美好。

亲爱的您，这里是LearingYard学苑！

今天小编为您带来“越览（128）——精读期刊论文

《基于共词网络的群智知识挖掘方法

——在应急决策中应用》的

2方法原理（1）”。

欢迎您的访问！

Share interest, spread happiness,

increase knowledge, and leave beautiful.

Dear, this is the LearingYard Academy!

Today, the editor brings the

＂Yue Lan (128)：Intensive reading of the journal article

＇Crowd intelligence knowledge mining method

based on co-word network– application

in emergency decision-making’

2 Principle of the method(1)".

Welcome to visit!

一、内容摘要（Summary of Content）

本期推文将从思维导图、精读内容、知识补充三个方面介绍精读复刻论文《基于共词网络的群智知识挖掘方法——在应急决策中应用》的2方法原理（1）。

This issue will introduce the principle of the method of the intensively read replica paper "Crowd intelligence knowledge mining method based on co-word network – application in emergency decision-making" in terms of mind maps, intensively read content, and knowledge supplementation.

二、思维导图（Mind mapping）

三、精读内容（Intensive reading content）

（一）方法框架（Methodology Framework）

本文提出一种基于社会网络行为大数据的群智知识挖掘方法，通过交互数据衡量数据价值，利用共词分析生成群智知识，并结合社会网络分析提取应急决策属性参数。最终，综合专家偏好构建动态决策支持方法。本文的方法及其应用过程整体框架如下图所示：

This paper proposes a method for crowd intelligence knowledge mining based on social network behavior big data. It measures data value through interaction data, generates crowd intelligence knowledge using co-word analysis, and extracts emergency decision attribute parameters in combination with social network analysis. Finally, a dynamic decision support method is constructed based on expert preferences. The overall framework of this method and its application process is shown in the figure below:

（二）群智知识的获取与处理（Acquisition and processing of crowd intelligence knowledge）

1. 数据获取（Acquisition and processing of crowd intelligence knowledge）

本文选择新浪微博作为群智知识的获取渠道，利用Python爬取用户发布、转发、评论、点赞等行为数据。通过相关关键词筛选与特大公共安全事件（如新冠肺炎疫情）相关的信息，以满足应急决策的及时性需求。

This paper selects Sina Weibo as a channel for acquiring crowd intelligence knowledge, and uses Python to crawl user posting, forwarding, commenting, liking and other behavioral data. Information related to major public safety events (such as the COVID-19 epidemic) is filtered through relevant keywords to meet the timeliness requirements of emergency decision-making.

2. UGCS特征表示（UGCS feature representation）

本文借鉴大数据分析的逻辑，利用TF-IDF算法对微博博文中的用户生成内容（UGCs）进行关键词提取。由于微博中存在大量生活化表达，简单的词频统计难以准确衡量术语的重要性。TF-IDF通过为术语分配分数，有效区分高频词和关键词，从而提取出具有代表性的特征关键词，实现UGCs的特征表示。

This paper draws on the logic of big data analysis and uses the TF-IDF algorithm to extract keywords from user-generated content (UGCs) in Weibo posts. Since there are a large number of daily expressions in Weibo, simple word frequency statistics are difficult to accurately measure the importance of terms. TF-IDF effectively distinguishes high-frequency words and keywords by assigning scores to terms, thereby extracting representative feature keywords and realizing the feature representation of UGCs.

（三）基于共词网络的群智知识可视化分析（Visual analysis of crowd intelligence knowledge based on co-word network）

本节结合社会网络分析与关键词共现分析，构建共词网络以实现危机事件下的公众群智知识可视化。关键词共现分析通过识别在同一博文中共同出现的关键词，揭示研究主题和语义关联。

This section combines social network analysis with keyword co-occurrence analysis to construct a co-word network to visualize the public wisdom knowledge under crisis events. Keyword co-occurrence analysis reveals research topics and semantic associations by identifying keywords that co-occur in the same blog post.

不同于传统方法仅基于词频计算网络边权，本文改进了共词网络构建方式，考虑了博文在社会网络中的影响力。通过专家打分法为转发、评论、点赞等交互行为设定权重，以更准确地衡量UGCs的综合影响力，从而优化共现强度的计算。

Different from the traditional method of calculating network edge weights based only on word frequency, this paper improves the co-word network construction method and considers the influence of blog posts in social networks. By using the expert scoring method, weights are set for interactive behaviors such as forwarding, commenting, and liking to more accurately measure the comprehensive influence of UGCs, thereby optimizing the calculation of co-occurrence intensity.

接着通过引入UGCs影响力和Word2Vec计算的语义相似度，改进了共词网络的共现关系描述。结合内部共现信息和外部语义知识，提升了群智知识挖掘和社区检测的准确性。

Then, by introducing the influence of UGCs and the semantic similarity calculated by Word2Vec, the co-occurrence relationship description of the co-word network was improved. Combining internal co-occurrence information and external semantic knowledge, the accuracy of crowd intelligence knowledge mining and community detection was improved.

四、知识补充（Knowledge Supplement）

UGCs（User-Generated Content，用户生成内容）是指由平台的用户而非专业机构或企业创建的任何形式的内容。通常，这些内容通过社交媒体、在线平台、论坛、博客等渠道发布，并且在互联网上广泛传播。

UGCs (User-Generated Content) refers to any form of content created by platform users rather than professional organizations or companies. Usually, these contents are published through social media, online platforms, forums, blogs and other channels, and widely circulated on the Internet.

在某些学术研究中，UGC被用来分析和衡量社交媒体或其他网络平台的影响力。例如，在博文或文章的影响力评估中，UGC的数量（如转发、评论、点赞）通常作为衡量内容受欢迎程度和社会影响力的重要指标。这些交互行为的权重（如转发可能比评论更能体现博文的影响力）通常会通过权重系数来量化，以便更准确地评估其综合影响力。

In some academic studies, UGC is used to analyze and measure the influence of social media or other online platforms. For example, in the impact assessment of blog posts or articles, the number of UGC (such as forwarding, comments, and likes) is usually used as an important indicator to measure the popularity and social influence of the content. The weight of these interactive behaviors (such as forwarding may better reflect the influence of a blog post than commenting) is usually quantified through weight coefficients to more accurately assess its comprehensive influence.

在一些研究中，UGC的影响力不仅与其内部的互动（例如博文的评论、点赞等）相关，还可能涉及外部的语义相似度信息（例如关键词的语义关系）。通过这种方式，研究人员希望更全面地理解内容在网络中的传播模式和其对用户行为的影响。

In some studies, the influence of UGC is not only related to its internal interactions (such as comments and likes on blog posts), but may also involve external semantic similarity information (such as the semantic relationship of keywords). In this way, researchers hope to have a more comprehensive understanding of the content dissemination pattern in the network and its impact on user behavior.

今天的分享就到这里了。

如果您对文章有独特的想法，

欢迎给我们留言，让我们相约明天。

祝您今天过得开心快乐！

That's all for today's sharing.

If you have a unique idea about the article,

please leave us a message,

and let us meet tomorrow.

I wish you a nice day!

文案|yyz

排版|yyz

审核|hzy