CN103136331A - Micro blog network opinion leader identification method - Google Patents

Micro blog network opinion leader identification method Download PDF

Info

Publication number
CN103136331A
CN103136331A CN2013100278084A CN201310027808A CN103136331A CN 103136331 A CN103136331 A CN 103136331A CN 2013100278084 A CN2013100278084 A CN 2013100278084A CN 201310027808 A CN201310027808 A CN 201310027808A CN 103136331 A CN103136331 A CN 103136331A
Authority
CN
China
Prior art keywords
node
ui
nodes
weights
network
Prior art date
Application number
CN2013100278084A
Other languages
Chinese (zh)
Inventor
蔡霖
蔡皖东
彭冬
Original Assignee
西北工业大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 西北工业大学 filed Critical 西北工业大学
Priority to CN2013100278084A priority Critical patent/CN103136331A/en
Publication of CN103136331A publication Critical patent/CN103136331A/en

Links

Abstract

The invention discloses a micro blog network opinion leader identification method and is used for solving the technical problem of poor recall rate of the existing opinion leader identification method. The technical scheme includes storing network topology information collected from the internet into a database by using a web crawler tool; constructing a directed network diagram G=(V, E); calculating an effective fan collection Ef (u); calculating a node weight IRL (ui) generated by a link relationship; calculating a node weight IRTR (ui) generated by a node interactive relationship; calculating a node comprehensive weight IR (ui); and calculating comprehensive weights of all nodes in the network diagram, performing sequencing in a descending order according to the comprehensive weights, and selecting n nodes with larger comprehensive weights as candidates of opinion leaders. The fan number possessed by the nodes, factors such as the node link relationship and the interactive relationship are considered when the node weights are calculated, so that the recall rate and the accuracy are improved. Through detection, the return rate is improved from 81.7-88.5% of the background technology to above 89.3%, and the accuracy is improved from 84.7-90.4% of the background technology to above 91.7%.

Description

微博网络意见领袖识别方法 Microblogging network of opinion leaders identification method

技术领域 FIELD

[0001] 本发明涉及一种识别方法,具体涉及一种微博网络意见领袖识别方法。 [0001] The present invention relates to a method of identification, more particularly to a micro-blog network identification method opinion leaders.

背景技术 Background technique

[0002] 随着Web2.0技术的发展,互联网中出现了一些新型的网络应用,如社交网络、微博网络等,在信息传播和人际关系承载方面,显现出越来越大的价值和影响力。 [0002] With the development of Web2.0 technology, the Internet there have been some new network applications such as social networking, microblogging networks, in terms of information dissemination and interpersonal bearer, show increasing value and impact force.

[0003] 社交网络作为现实社交网络在互联网上的扩展,旨在帮助人们更加有效地建立和维持人际关系网络。 [0003] as a real social network of social networks on the Internet expansion, designed to help people more effectively establish and maintain interpersonal networks. 与以聚合信息为特点的网站不同,社交网络以聚合人群为特点,人们通过社交网络可以建立和维持自己的朋友圈子,成为一种新型的个人社交方式和信息交流平台,借助于朋友口碑相传的信息传播模式,加速了信息的传播。 And aggregate information as to the characteristics of different sites, social networks to aggregate population characteristics, one can establish and maintain their own circle of friends through social networks, become a new kind of personal and social way information exchange platform, by means of word of mouth of friends information dissemination mode, accelerate the dissemination of information.

[0004] 微博网络(Micro-Blogging Network)也是一种社交网络,用户可以通过浏览器、手机、即时通信软件等多种渠道发布140字以内的信息,这种即时性、碎片化、聚合性的信息传播特性受到广大用户的欢迎,国内新浪微博的注册用户已超过3亿人。 [0004] microblogging network (Micro-Blogging Network) is a social network, users can browse, cell phones, instant messaging software and other channels Ads less than 140 words, this immediacy, fragmentation, polymeric the information propagation characteristics by the majority of users are welcome, domestic Sina microblogging registered users has more than 300 million people.

[0005] 在网络信息传播过程中,意见领袖发挥了重要的作用。 [0005] In the process of information dissemination network, opinion leaders play an important role. 局部意见在意见领袖的引导和影响下演化为网络舆论。 Local opinion evolved into a network under the guidance of public opinion and influence opinion leaders. 意见领袖又称舆论领袖,是指在人际传播网络中经常为他人提供信息并施加影响的“活跃分子”,他们在大众传播效果的形成过程中起着重要的中介或过滤作用,由他们将信息扩散给受众,形成信息传递的两级传播。 Opinion leaders also called opinion leaders, often refers to the provision of information in interpersonal communication networks for others and influence of the "activists", they play an important intermediary role in the formation or filtration dissemination of results of the general population, by their information diffusion to the audience, two spread information transmission form. 随着网络舆论影响力的不断加大,人们对网络意见领袖的研究也在不断地深入。 With the increasing influence of the network of public opinion, people study the network of opinion leaders are constantly deepened.

[0006] 统计数据显示,网络中的大部分用户不经常参与信息的制造与传播,他们做出的决定往往跟随意见领袖。 [0006] statistics show that the majority of network users do not regularly involved in the manufacture and dissemination of information, the decision they make tends to follow the opinion leaders. 有效地识别网络意见领袖,通过意见领袖发表引导性信息来影响所在网络用户而非直接说服他们,可以有效地触发整个网络或社会的影响力,对于推动信息传播,提高广告效应具有重要的现实意义。 Effectively identify network opinion leaders, information guide published by opinion leaders to influence the users on your network rather than directly to convince them that can effectively trigger the influence of the entire network or community, to promote the dissemination of information to improve the advertising effect has important practical significance .

[0007] 微博网络是一种复杂网络,国内外通常采用复杂网络理论对这类复杂网络进行建模分析,以揭示复杂网络的内在特性和科学规律。 [0007] microblogging network is a complex network, usually at home and abroad for this type of complex network theory to model complex network analysis to reveal the intrinsic properties of complex networks and scientific laws. 根据复杂网络理论,可将微博网络抽象成一种有向网络图,每个用户构成网络中的节点,用户之间关系构成节点之间的边,由于每个用户拥有的朋友和粉丝数量不同,因此各个节点具有不同的权值,节点权值越大,说明该节点的影响力越大,成为意见领袖的可能性也就越大。 The complex network theory, Twitter network abstraction may be directed into a network diagram, each of the user nodes in the network configuration, user configuration relationship between the edges between the nodes, since each user has a different number of fans and friends, Therefore, each node has a different weight, the greater the node weights, the larger the influence of the node, the possibility of becoming opinion leaders greater. 因此意见领袖识别问题可归结为如何计算节点权值问题。 Therefore, opinion leaders recognize the problem boils down to how to calculate the node weight problem. 通过建立有向网络图模型,分析节点之间结构关系,计算每个节点的权值,节点权值越大,成为意见领袖的可能性就越大。 There are structural relationship between the network graph model to analyze the nodes, each node to calculate the weight, the greater the node weights, the possibility of becoming opinion leaders through the establishment of the greater.

[0008] 文献I “Discovering Important Bloggers based on Analyzing BlogThreads [WWW, Chiba, Japan, May 10-14,2005] ”提出了一种基于帖子内容分析的博客重要用户分析方法ThreadRank,该方法通过分析大量的博客内容来判断其用户的重要性,需要耗费大量的时间用于内容清理和分析,效率较低。 [0008] Document I "Discovering Important Bloggers based on Analyzing BlogThreads [WWW, Chiba, Japan, May 10-14,2005]" raised important blog user based on analysis ThreadRank post content analysis, the method by analyzing a large number of blog content to determine the importance of its users and requires a lot of time to clean up and analysis of the content, the lower the efficiency.

[0009]文献 2“ Identifying Opinion Leaders in the Blogosphere [CIKM, pp.971-974,2007] ”提出了一种意见领袖识别方法Inf luenceRank,该方法根据与其他博客相比较来判断用户的重要性,以及这些用户对整个网络所做的贡献来计算用户权值,并采用余弦定理计算不同博客实体的相似性,复杂性较高,开销大。 [0009] Document 2 "Identifying Opinion Leaders in the Blogosphere [CIKM, pp.971-974,2007]" opinion leaders proposed a recognition method Inf luenceRank, the method according to the importance compared to other blog user to judge, and the contribution made by these users to the entire network of users to calculate the weights, and the use of similar nature, complexity higher law of cosines to calculate different blog entities, large overhead.

[0010]文献 3 “TwitterRank:Finding topic-sensitive Inuential Twitterers [WSDM,2010] ”提出了一种Twitter网络节点计算方法TwitterRank,该方法根据Twitter中的用户关系、粉丝与关注者之间的分布以及在信息传播的过程中各种用户群体所起到的作用进行权重计算,该算法主要基于话题进行分析,召回率不高。 [0010] Document 3 "TwitterRank: Finding topic-sensitive Inuential Twitterers [WSDM, 2010]" the calculation method TwitterRank Twitter one kind of a network node, according to a user relationship Twitter distribution and the method, between the fans and followers the process of information dissemination in the various user groups carried out the role of weight calculation, the algorithm is based primarily on the topic of the analysis, the recall rate is not high.

发明内容 SUMMARY

[0011] 为了克服现有意见领袖识别方法召回率差的不足,本发明提供一种微博网络意见领袖识别方法。 [0011] In order to overcome the existing opinion leaders difference recognition rate is less than a recall, the present invention provides a micro-blog network identification method opinion leaders. 该方法通过节点权重来标识节点影响力和重要性,节点权值越大,成为意见领袖的可能性就越大。 In this method, the possibility of node weights greater influence and importance of identifying a node, the node weights, opinion leaders will be. 在计算节点权重时,考虑到节点拥有的粉丝数量以及节点链接关系和交互关系等多方面因素,可以提闻召回率,同时提闻准确率。 In calculating the node weights, taking into account various factors and the number of nodes fans link relationships and interactions peers have, mention may smell recall, while mentioning the smell accuracy.

[0012] 本发明解决其技术问题所采用的技术方案是:一种微博网络意见领袖识别方法,其特点是包括以下步骤: [0012] aspect of the present invention to solve the technical problem is: A network Twitter opinion leaders identification method, characterized by comprising the steps of:

[0013] 步骤一、利用网络爬虫工具,从互联网中采集实际的微博网络数据,提取其中的网络拓扑信息存入数据库待处理。 [0013] Step a, using a web crawler tools, Twitter collected actual data from the Internet network, extract the network topology information stored in the database to be processed.

[0014] 步骤二、构建微博有向网络图 [0014] Step two, the network constructed with a micro-blog FIG.

[0015] G = (E, V) [0015] G = (E, V)

[0016] 式中,E表示节点关系集合,V表示节点集合。 [0016] In the formula, E represents a set of node relationship, V represents the set of nodes.

[0017] 步骤三、计算有效粉丝集合Ef (U) [0017] Step three, calculate the effective fan set Ef (U)

[0018] Ef(u) = {vI V e Follower (U) Λ Response (U)>δ} [0018] Ef (u) = {vI V e Follower (U) Λ Response (U)> δ}

[0019] 式中,δ是非负常数阈值,表示节点U的粉丝节点V对节点u的反馈程度门限,超过该阈值且属于节点u的粉丝才能算作有效粉丝。 [0019] wherein, [delta] non-negative constant threshold value, the nodes representing node U fans V u feedback node degree threshold, the threshold is exceeded, and node u belongs to the fans can be counted as valid fans.

[0020] 步骤四、计算由链接关系所产生的节点权值IRL(Ui) [0020] Step 4 is calculated by the link relation node weight values ​​generated IRL (Ui)

Figure CN103136331AD00041

[0022] 式中,Follower(Ui)为节点Ui的所有粉丝集合,L(Uj)为节点Uj的粉丝数目,σ是介于O和I的阻尼系数,N为网络图中的总节点数。 [0022] In the formula, Follower (Ui) of all the fans node set Ui, L (Uj) is the number of fans nodes Uj, σ is between O and I damping coefficient, N is the total number of nodes in the network of FIG.

[0023] 步骤五、计算由节点交互关系所产生的节点权值IRTR(Ui) [0023] Step 5 is calculated by the node weights interactions node values ​​generated IRTR (Ui)

Figure CN103136331AD00042

[0025] 式中,Tweet(Ui)为节点Ui帖子集合,A表示所有具有交互情况的帖子集|A|是A的集合,Ns(Uj)是节点Uj针对帖子t」,的响应次数,Nli (Uj)为响应平均值,Response包括用户转帖、回帖、评论和收藏。 [0025] where, Tweet (Ui) for the node Ui post collection, A represents all posts set with interactive case | A | is set A, Ns (Uj) is the node Uj for posts t ", the response times, Nli (Uj) is the mean response, response, including the user posted, replies, comments and collections.

[0026] 步骤六、计算节点综合权值IR(Ui) [0026] Step 6 Synthesis weight computing node values ​​IR (Ui)

[0027] IR (Ui) = (1- β ) X IRL (Ui) + β X IRTR (Ui) [0027] IR (Ui) = (1- β) X IRL (Ui) + β X IRTR (Ui)

[0028] 式中,参数β (β e [0,1])决定链接关系和节点交互关系两个因子在节点权值计算中所处的地位;当β较小时,节点权值由链接关系决定,特别当β = O时则完全由链接关系来计算权值。 [0028] In the formula, the parameter β (β e [0,1]) to determine the connection relationship and the position in which the two node weight value calculation factor interactions node; when beta] is small, the node weight is determined by the link relation , particularly when β = O is calculated by the link relation weights completely. [0029] 步骤七、计算网络图中所有节点的综合权值,并按综合权值由大到小排序,选取综合权值较大的η个节点,作为意见领袖的候选对象。 [0029] Step 7 in FIG integrated computing network weights of all nodes, an integrated press weights descending order, selecting a larger integrated weights η nodes as candidates opinion leaders.

[0030] 本发明的有益效果是:由于通过节点权重来标识节点影响力和重要性,节点权值越大,成为意见领袖的可能性就越大。 [0030] the advantages are: the adoption of a node identifies the node weights greater influence and importance of the node weights, the possibility of becoming opinion leaders will be. 在计算节点权重时,考虑到节点拥有的粉丝数量以及节点链接关系和交互关系等多方面因素,提高了召回率,同时提高了准确率。 In calculating the node weights, taking into account various factors and the number of nodes fans link relationships and interactions peers have, to improve the recall rate, while improving accuracy. 经检测,召回率由背景技术的81.7〜88.5%提高到89.3%以上,准确率由背景技术的84.7〜90.4%提高到91.7%以上。 After testing, 81.7~88.5% recall rate is increased to more than BACKGROUND 89.3%, accuracy rate Background 84.7~90.4% to 91.7% increase.

[0031] 下面结合附图和实施例对本发明作详细说明。 Drawings and embodiments of the present invention will be described in detail [0031] below in conjunction.

附图说明 BRIEF DESCRIPTION

[0032] 图1是本发明微博网络意见领袖识别方法的流程图。 [0032] FIG. 1 is a flowchart illustrating the opinion leaders Twitter network identification method of the present invention.

具体实施方式 Detailed ways

[0033] 参照图1。 [0033] Referring to FIG. 本发明微博网络意见领袖识别方法具体步骤如下: Twitter opinion leaders network identification method of the present invention specific steps are as follows:

[0034] 1.获取微博网络数据:利用网络爬虫工具,从互联网中采集实际的微博网络数据,提取其中的节点、连接等网络拓扑信息存入数据库待处理。 [0034] 1. Obtain data network Twitter: web crawler using tools, Twitter collected actual data from the Internet network, wherein the node extracted, and the like connected to the network topology information stored in the database to be processed.

[0035] 2.构建微博有向网络图: [0035] 2. Construction of the vector graph Twitter:

[0036] G = (Ε, V) [0036] G = (Ε, V)

[0037] 式中,E表示节点关系集合,V表示节点集合。 [0037] In the formula, E represents a set of node relationship, V represents the set of nodes.

[0038] 3.计算有效粉丝集合Ef (U): [0038] 3. Compute the effective fan set Ef (U):

[0039] Ef (u) = {ν IV e Follower (U) Λ Response (U) > δ } (I) [0039] Ef (u) = {ν IV e Follower (U) Λ Response (U)> δ} (I)

[0040] 式中,δ是非负常数阈值,表示节点U的粉丝节点V对节点u的反馈程度门限,超过该阈值且属于节点u的粉丝才能算作有效粉丝。 [0040] wherein, [delta] non-negative constant threshold value, the nodes representing node U fans V u feedback node degree threshold, the threshold is exceeded, and node u belongs to the fans can be counted as valid fans.

[0041] 4.计算由链接关系所产生的节点权值IRL(Ui): [0041] 4. Calculate the weight of the link relation node generated IRL (Ui):

[0042] [0042]

Figure CN103136331AD00051

[0043] 式中,Follower(Ui)为节点Ui的所有粉丝集合,L(Uj)为节点Uj的粉丝数目,σ是介于O和I的阻尼系数,N为网络图中的总节点数。 [0043] In the formula, Follower (Ui) of all the fans node set Ui, L (Uj) is the number of fans nodes Uj, σ is between O and I damping coefficient, N is the total number of nodes in the network of FIG.

[0044] 5.计算由节点交互关系所产生的节点权值IRTR(Ui): [0044] 5. compute node weights IRTR (Ui) produced by the interactions of nodes:

[0045] [0045]

Figure CN103136331AD00052

[0046] 式中,Tweet(Ui)为节点Ui帖子集合,A表示所有具有交互情况的帖子集|A|是A的集合,Ns(Uj)是节点Uj针对帖子t」,的响应次数,Nli (Uj)为响应平均值,Response包括用户转帖、回帖、评论和收藏。 [0046] where, Tweet (Ui) for the node Ui post collection, A represents all posts set with interactive case | A | is set A, Ns (Uj) is the node Uj for posts t ", the response times, Nli (Uj) is the mean response, response, including the user posted, replies, comments and collections.

[0047] 6.计算节点综合权值IR(Ui): [0047] 6. The integrated computing node weights IR (Ui):

[0048] IR (Uj) = (1- β ) X IRL (Ui) + β X IRTR (Ui) (4) [0048] IR (Uj) = (1- β) X IRL (Ui) + β X IRTR (Ui) (4)

[0049] 式中,参数β (β e [0,1])决定链接关系和节点交互关系两个因子在节点权值计算中所处的地位;当β较小时,节点权值主要由链接关系决定,特别当β =O时则完全由链接关系来计算权值。 [0049] In the formula, the parameter β (β e [0,1]) and the determined position values ​​of the two factors in which the node weight calculation link relation interactions node; when beta] is small, the node link relation mainly by weight decision, particularly when β = O is calculated entirely from the weight of the link relation.

[0050] 7.计算网络图中所有节点的综合权值,并按综合权值由大到小排序,选取综合权值较大的η个节点,作为意见领袖的候选对象。 [0050] Integrated weights of all nodes in FIG. 7. The computing network, in accordance descending order of weighting values, selecting a larger integrated weights η nodes as candidates opinion leaders.

[0051] 本发明从计算效率和精确度两个方面改进了现有方法的不足。 [0051] The present invention improves the deficiencies of prior methods from both computational efficiency and accuracy. 首先,通过定义有效粉丝集合,将没有或拥有少量粉丝的节点排除掉,他们成为意见领袖的可能性极小,因为意见领袖或高权值节点必然拥有大量粉丝,这样就可大幅度减小网络图规模,有利于提高计算效率。 First, by defining a set of valid fans will not have a node or a small number of fans excluded, they become opinion leaders the possibility of very small, because opinion leaders or high node weights necessarily have a lot of fans, so you can greatly reduce network Figure scale, will help improve computational efficiency. 其次,在计算节点权值时,不仅考虑了由粉丝产生的链接关系,还考虑了帖子的发布、转发、回复以及收藏等所产生的节点交互关系,因此提高了计算精确度。 Secondly, in calculating the node weights, considering not only the link relations generated by the fans, but also takes into account the post release, forward, reply, and collection and other node interactions generated, thereby improving the accuracy of the calculation.

[0052] 本发明与现有方法对比实验结果如表I所示。 [0052] The present invention and comparative experimental results shown in Table I as the conventional method.

[0053] 表I各种节点权值计算方法的召回率、准确率及平均节点处理时间对照 [0053] Recall Table I calculated various nodes of the weights, the mean accuracy and processing time of control nodes

[0054] [0054]

Figure CN103136331AD00061

[0055] 该实验是以处理10万个微博网络节点为基准测试的。 [0055] The experiment is based on processing 100,000 microblogging network node benchmark. 从表I中可以看出,基于出度、入度/出度结合等方法虽然计算效率较高,但准确率和召回率很低;文献1、文献2和文献3提出的方法虽然具有较高的准确率和召回率,但计算效率比较低;而本发明不仅具有较高的计算效率,并且还具有较高的准确率和召回率。 As can be seen from Table I, the degree of penetration / binding of other methods based on the high computational efficiency, although, but the precision and recall rate is low; Document 1, Document 2, and Document 3, a method is proposed, although with higher precision and recall, but the calculation efficiency is relatively low; the present invention not only has higher efficiency, and also has a high precision and recall.

Claims (1)

1.一种微博网络意见领袖识别方法,其特征在于包括以下步骤: 步骤一、利用网络爬虫工具,从互联网中采集实际的微博网络数据,提取其中的网络拓扑信息存入数据库待处理; 步骤二、构建微博有向网络图G =(E7V) 式中,E表示节点关系集合,V表示节点集合; 步骤三、计算有效粉丝集合Ef(U) Ef(u) = {vI V e Follower(u) Λ Response(u)> δ} 式中,S是非负常数阈值,表示节点U的粉丝节点V对节点U的反馈程度门限,超过该阈值且属于节点U的粉丝才能算作有效粉丝; 步骤四、计算由链接关系所产生的节点权值IRL(Ui) An opinion leaders Twitter network identification method, comprising the steps of: a step, using a web crawler tools, Twitter collected actual data from the Internet network, extract the network topology information stored in the database to be processed; step two, constructed Twitter the vector graph G = (E7V) formula, E represents the node set of relationships, V represents the set of nodes; step three, calculate the effective fan set Ef (U) Ef (u) = {vI V e Follower (u) Λ Response (u)> δ} where, S is a nonnegative constant threshold value, it indicates the node U fans nodes V feedback level of the gate node U limit, exceeds the threshold and belonging to the node U fans can be counted as valid fan; step four, the compute node weights IRL (Ui) produced by the link relation
Figure CN103136331AC00021
式中,Follower(Ui)为节点Ui的所有粉丝集合,L(Uj)为节点Uj的粉丝数目,σ是介于O和I的阻尼系数,N为网络图中的总节点数; 步骤五、计算由节点交互关系所产生的节点权值IRTR(Ui) Wherein, Follower (Ui) of all the fans node set Ui, L (Uj) is the number of fans nodes Uj, [sigma] is the damping coefficient between O and I, N is the total number of nodes in the network map; Step 5 node weights calculated by the interactions node values ​​generated IRTR (Ui)
Figure CN103136331AC00022
Ui帖子集合,A表示所有具有交互情况的帖子集|A|是A的集合,Ns(Uj)是节点Uj针对帖子tj,的响应次数,Nli (Uj)为响应平均值,Response包括用户转帖、回帖、评论和收藏; 步骤六、计算节点综合权值IR(Ui) IR (Ui) = (1- β ) X IRL (Ui) + β X IRTR (Ui) 式中,参数β (β e [0,1])决定链接关系和节点交互关系两个因子在节点权值计算中所处的地位;当β较小时,节点权值由链接关系决定,特别当β = O时则完全由链接关系来计算权值; 步骤七、计算网络图中所有节点的综合权值,并按综合权值由大到小排序,选取综合权值较大的η个节点,作为意见领袖的候选对象。 Post set Ui, A represents a set of all posts in the case of interaction with | A | is set A, Ns (Uj) is a node for a message Uj tj, response times, Nli (Uj) is the average response, Response including user Posts , Replies, comments and collections; step 6 integrated computing node weights IR (Ui) IR (Ui) = (1- β) X IRL (Ui) + β X IRTR (Ui) wherein, the parameter β (β e [ 0,1]) to determine the connection status of the relationship between two factors and interactions in the node value calculation in which the node weight; when beta] is small, the node weight is determined by the link relation, particularly when β = entirely by the link when the relation O calculating weights; step 7 integrated weights of all nodes in FIG computing network, in accordance descending order of weighting values, selecting a larger integrated weights η nodes as candidates opinion leaders.
CN2013100278084A 2013-01-18 2013-01-18 Micro blog network opinion leader identification method CN103136331A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2013100278084A CN103136331A (en) 2013-01-18 2013-01-18 Micro blog network opinion leader identification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2013100278084A CN103136331A (en) 2013-01-18 2013-01-18 Micro blog network opinion leader identification method

Publications (1)

Publication Number Publication Date
CN103136331A true CN103136331A (en) 2013-06-05

Family

ID=48496157

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2013100278084A CN103136331A (en) 2013-01-18 2013-01-18 Micro blog network opinion leader identification method

Country Status (1)

Country Link
CN (1) CN103136331A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105723402A (en) * 2013-10-25 2016-06-29 西斯摩斯公司 Systems and methods for determining influencers in a social data network
CN105959368A (en) * 2016-04-29 2016-09-21 成都信息工程大学 Social cloud hot spot resource prediction and disposition method
CN106055627A (en) * 2016-05-27 2016-10-26 西安电子科技大学 Recognition method of key nodes of social network in topic field

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102214212A (en) * 2011-05-20 2011-10-12 西北工业大学 Method for ordering microblog network node weights based on multi-link
CN102662956A (en) * 2012-03-05 2012-09-12 西北工业大学 Method for identifying opinion leaders in social network based on topic link behaviors of users

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102214212A (en) * 2011-05-20 2011-10-12 西北工业大学 Method for ordering microblog network node weights based on multi-link
CN102662956A (en) * 2012-03-05 2012-09-12 西北工业大学 Method for identifying opinion leaders in social network based on topic link behaviors of users

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105723402A (en) * 2013-10-25 2016-06-29 西斯摩斯公司 Systems and methods for determining influencers in a social data network
CN105959368A (en) * 2016-04-29 2016-09-21 成都信息工程大学 Social cloud hot spot resource prediction and disposition method
CN105959368B (en) * 2016-04-29 2019-04-02 成都信息工程大学 A kind of method of social activity cloud hot point resource prediction and deployment
CN106055627A (en) * 2016-05-27 2016-10-26 西安电子科技大学 Recognition method of key nodes of social network in topic field
CN106055627B (en) * 2016-05-27 2019-06-18 西安电子科技大学 The recognition methods of social networks key node in topic field

Similar Documents

Publication Publication Date Title
Scellato et al. Socio-spatial properties of online location-based social networks
US10069931B2 (en) Method and system for pushing recommended friend to user of social network
Nguyen et al. Real-time event detection for online behavioral analysis of big social data
CN102663101A (en) Sina microblog-based user grade sequencing algorithm
CN103942340A (en) Microblog user interest recognizing method based on text mining
CN100583804C (en) Method and system for processing social network expert information based on expert value propagation algorithm
CN102346766A (en) Method and device for detecting network hot topics found based on maximal clique
Dong et al. An experimental study of large-scale mobile social network
CN102571484A (en) Method for detecting and finding online water army
CN103428164B (en) User social network relationship division method and system
CN104462213A (en) User behavior analysis method and system based on big data
TW201342280A (en) Network virtual users risk control method and system
CN103617279A (en) Method for achieving microblog information spreading influence assessment model on basis of Pagerank method
CN103024017A (en) Method for distinguishing important goals and community groups of social network
CN102662956A (en) Method for identifying opinion leaders in social network based on topic link behaviors of users
CN103544188B (en) Method and device user preferences pushing mobile Internet content
CN102117325A (en) Method for predicting dynamic social network user behaviors
Zhaoyun et al. Mining topical influencers based on the multi-relational network in micro-blogging sites
CN102347917B (en) Contact semantic grouping method for network message communication
CN102200987A (en) Method and system for searching sock puppet identification number based on behavioural analysis of user identification numbers
CN103116605B (en) A real-time detection method and system for hot microblogging event-based monitoring subnet
CN103092828B (en) Based on semantic text analysis and semantic networks similarity metrics
Jabeur et al. Active microbloggers: identifying influencers, leaders and discussers in microblogging networks
CN103745000B (en) A method for detecting a hot topic of Chinese micro-blog
CN102571485B (en) Method for identifying robot user on micro-blog platform

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)