LIN et al 2023a - Revision history

Rimni at 09:54, 21 December 2023

2023-12-21T09:54:11Z

Rimni at 09:53, 21 December 2023

2023-12-21T09:53:16Z

Rimni at 09:49, 21 December 2023

2023-12-21T09:49:40Z

Rimni at 09:26, 21 December 2023

2023-12-21T09:26:10Z

Rimni at 09:25, 21 December 2023

2023-12-21T09:25:14Z

Rimni at 09:24, 21 December 2023

2023-12-21T09:24:23Z

Rimni at 09:16, 21 December 2023

2023-12-21T09:16:01Z

Rimni at 15:17, 20 December 2023

2023-12-20T15:17:56Z

Rimni at 15:17, 20 December 2023

2023-12-20T15:17:03Z

Rimni: /* 4. Analysis of key person in persistent topic with online Forum */

2023-12-20T15:15:54Z

‎4. Analysis of key person in persistent topic with online Forum

@@ Line 408: / Line 408: @@
 ===5.8 The Top N key persons in PTSN===
-Extracting the Top <math display="inline"> N </math> key persons in <math>PTSN</math> is achieved through a ranking nodes process based on the importance degree. The algorithm sorts the nodes according to the <math> SWNP</math>, and then modifies the list using the emotional attributes. The comparing algorithms mainly used are <math> IDC </math> (Indegree Prestige Centrality), <math>ODC  </math> (Outdegree Prestige Centrality) and <math>PR</math> (PageRank). <math> IDC </math> is based on the indegree number, so it takes into account the number of members that are adjacent to a particular member of the community, as follows: <math display="inline">IDC(x) =i(x)/(m−1)</math>, where <math display="inline"> m </math> is the number of nodes in the network, and <math>i(x)</math> is the number of members from the first level neighborhood that are adjacent to <math display="inline">  x</math>. In other words, more prominent people receive more nominations from members of the community. <math>ODC  </math> takes into account the outdegree number of the member <math display="inline">  x</math> for edges that are directed to the given node, as follows: <math display="inline">ODC(x) =o(x)/(m−1)</math>, where <math display="inline">o(x)</math> is the number of the first level neighbors to <math display="inline">  x</math>. On the other hand, users who have low outdegree centrality are not very open to the external world and do not communicate with many members. <math>ODC  </math> and <math> IDC </math> are the simplest and most intuitive measures that can be used in network analysis. Google uses <math>PR</math> to rank the pages in its search engine to measure the importance of a particular page to the others. [[#tab-2|Table 2]] gives the top 10 important nodes using different methods in the <math>No.16\, PTSN</math> with 3450 nodes.
+Extracting the Top <math display="inline"> N </math> key persons in <math>PTSN</math> is achieved through a ranking nodes process based on the importance degree. The algorithm sorts the nodes according to the <math> SWNP</math>, and then modifies the list using the emotional attributes. The comparing algorithms mainly used are <math> IDC </math> (Indegree Prestige Centrality), <math>ODC  </math> (Outdegree Prestige Centrality) and <math>PR</math> (PageRank). <math> IDC </math> is based on the indegree number, so it takes into account the number of members that are adjacent to a particular member of the community, as follows: <math display="inline">IDC(x) =i(x)/(m-1)</math>, where <math display="inline"> m </math> is the number of nodes in the network, and <math>i(x)</math> is the number of members from the first level neighborhood that are adjacent to <math display="inline">  x</math>. In other words, more prominent people receive more nominations from members of the community. <math>ODC </math> takes into account the outdegree number of the member <math display="inline">  x</math> for edges that are directed to the given node, as follows: <math display="inline">ODC(x) =o(x)/(m-1)</math>, where <math display="inline">o(x)</math> is the number of the first level neighbors to <math display="inline">  x</math>. On the other hand, users who have low outdegree centrality are not very open to the external world and do not communicate with many members. <math>ODC  </math> and <math> IDC </math> are the simplest and most intuitive measures that can be used in network analysis. Google uses <math>PR</math> to rank the pages in its search engine to measure the importance of a particular page to the others. [[#tab-2|Table 2]] gives the top 10 important nodes using different methods in the <math>No.16\, PTSN</math> with 3450 nodes.
 <div class="center" style="font-size: 75%;">'''Table 2'''. Top 10 users in ''No.18 PTSN''</div>

@@ Line 382: / Line 382: @@
 ===5.7 Distribution characteristics of SWNP===
-Experiments analyze the distribution characteristics of <math> SWNP</math> in <math>18\, PTSN</math>, and [[#img-3|Figure 3]] gives the average <math> SWNP</math> and their standard deviation in <math>No.16</math> and <math>No.18</math> <math>PTSN</math> with different <math>\varepsilon</math>. The average SWNP does not depend on <math>\varepsilon</math>, and it can be formally demonstrated that the SWNP equals approximately 1 in all cases. On the other hand, the standard deviation differs substantially depending on <math>\varepsilon</math>: the greater the <math>\varepsilon</math>, the greater the standard deviation. Namely, the <math> SWNP</math> value has increased disproportionately with bigger <math>\varepsilon</math>, which has been proven by the experimental data.
+Experiments analyze the distribution characteristics of <math> SWNP</math> in <math>18\, PTSN</math>, and [[#img-3|Figure 3]] gives the average <math> SWNP</math> and their standard deviation in <math>No.16</math> and <math>No.18\,PTSN</math> with different <math>\varepsilon</math>. The average SWNP does not depend on <math>\varepsilon</math>, and it can be formally demonstrated that the SWNP equals approximately 1 in all cases. On the other hand, the standard deviation differs substantially depending on <math>\varepsilon</math>: the greater the <math>\varepsilon</math>, the greater the standard deviation. Namely, the <math> SWNP</math> value has increased disproportionately with bigger <math>\varepsilon</math>, which has been proven by the experimental data.
 <div id='img-3'></div>
@@ Line 393: / Line 393: @@
-The distribution characteristics of <math> SWNP</math> are determined by its network topology structure; for example, the standard deviation variation tendency of <math>No.18</math> is more noticeable than <math>No.16</math>. This result indicates the greater difference of <math> SWNP</math> in <math>No.18 PTSN</math>, as there are a few nodes with ultra value. It can also be noted that the average <math> SWNP</math> over 81% of users is less than 1. This result means that only a few members exceed the average value that equals 1. This result also shows that the members’ <math> SWNP</math> difference increased for greater <math>\varepsilon</math>, and it is valid for all the 18 <math>PTSN</math>. The <math>No.18 PTSN</math> has the standard deviations of the most obvious change: while <math>\varepsilon =0.9</math>, fewer than 1% of users have a <math> SWNP >1</math>, and these users are clearly important. [[#img-4|Figure 4]] shows the percentage of users with <math> SWNP\ge 1</math> and <math> SWNP\ge 2</math> within <math>No.18</math> and <math>No.16 PTSN</math> in relation to <math>\varepsilon</math>.
+The distribution characteristics of <math> SWNP</math> are determined by its network topology structure; for example, the standard deviation variation tendency of <math>No.18</math> is more noticeable than <math>No.16</math>. This result indicates the greater difference of <math> SWNP</math> in <math>No.18\, PTSN</math>, as there are a few nodes with ultra value. It can also be noted that the average <math> SWNP</math> over 81% of users is less than 1. This result means that only a few members exceed the average value that equals 1. This result also shows that the members’ <math> SWNP</math> difference increased for greater <math>\varepsilon</math>, and it is valid for all the <math>18\,PTSN</math>. The <math>No.18 \, PTSN</math> has the standard deviations of the most obvious change: while <math>\varepsilon =0.9</math>, fewer than 1% of users have a <math> SWNP >1</math>, and these users are clearly important. [[#img-4|Figure 4]] shows the percentage of users with <math> SWNP\ge 1</math> and <math> SWNP\ge 2</math> within <math>No.18</math> and <math>No.16 \,PTSN</math> in relation to <math>\varepsilon</math>.
 <div id='img-4'></div>
@@ Line 404: / Line 404: @@
-It can be seen that the different <math>PTSN</math>s have the same <math> SWNP</math> distribution trend, with the <math> SWNP\ge 1</math> nodes decreasing and <math> SWNP\ge 2</math> nodes increasing. The average percentage of nodes with <math> SWNP\ge 2</math> is 4.7% in all the 18 <math>PTSN</math> (<math>No.16</math> with 7.54% and <math>No.18</math> with 0.57%). This conclusion can help us identify the important nodes in persistent topic social networks. The percent of <math> SWNP\ge 1</math> and <math> SWNP\ge 2</math> are 3.12% and 0.49% in <math>No.18 \,PTSN</math> while <math>\varepsilon =0.7</math>, so it can be assured that 3.12% users are active users and the 0.49% users are key person in this topic. In fact, the greater the <math>\varepsilon</math>, the more distinguishable the results, but the larger number of iterations directly influences the processing time. Generally, the parameter is determined by the different network scales, but the nodes with high <math> SWNP</math> values do not necessarily represent key persons, as the adjacent nodes may pass a lot of negative energy (if the commitment function is less than 0). Therefore, sentiment analysis is needed to actually identify the key persons.
+It can be seen that the different <math>PTSN</math>s have the same <math> SWNP</math> distribution trend, with the <math> SWNP\ge 1</math> nodes decreasing and <math> SWNP\ge 2</math> nodes increasing. The average percentage of nodes with <math> SWNP\ge 2</math> is 4.7% in all the <math>18\,PTSN</math> (<math>No.16</math> with 7.54% and <math>No.18</math> with 0.57%). This conclusion can help us identify the important nodes in persistent topic social networks. The percent of <math> SWNP\ge 1</math> and <math> SWNP\ge 2</math> are 3.12% and 0.49% in <math>No.18 \,PTSN</math> while <math>\varepsilon =0.7</math>, so it can be assured that 3.12% users are active users and the 0.49% users are key person in this topic. In fact, the greater the <math>\varepsilon</math>, the more distinguishable the results, but the larger number of iterations directly influences the processing time. Generally, the parameter is determined by the different network scales, but the nodes with high <math> SWNP</math> values do not necessarily represent key persons, as the adjacent nodes may pass a lot of negative energy (if the commitment function is less than 0). Therefore, sentiment analysis is needed to actually identify the key persons.
 ===5.8 The Top N key persons in PTSN===

@@ Line 367: / Line 367: @@
 ===5.6 Node position iterative data processing===
-The experiments revealed that the number of iterations necessary to calculate the node positions for all users in each <math>PTSN</math> depends on the value of the parameter <math>\varepsilon  </math> (Eq.(6)): the greater the value of <math>\varepsilon</math>, the greater the number of iterations ([[#img-2|Figure 2]]). Each node was initialized in <math>PTSN</math> with <math> SWNP=1</math> and the stop condition <math>\tau =0.00001</math>. The iterative processing of <math> SWNP</math> uses six different <math>\varepsilon (0.01, 0.1, 0.3, 0.5, 0.7, and 0.9)</math> for comparative analysis. Because the given 18 <math>PTSN</math>s have similar sizes, their tendencies are similar.
+The experiments revealed that the number of iterations necessary to calculate the node positions for all users in each <math>PTSN</math> depends on the value of the parameter <math>\varepsilon  </math> (Eq.(6)): the greater the value of <math>\varepsilon</math>, the greater the number of iterations ([[#img-2|Figure 2]]). Each node was initialized in <math>PTSN</math> with <math> SWNP=1</math> and the stop condition <math>\tau =0.00001</math>. The iterative processing of <math> SWNP</math> uses six different <math>\varepsilon (0.01, 0.1, 0.3, 0.5, 0.7, \, \hbox{and } 0.9)</math> for comparative analysis. Because the given 18 <math>PTSN</math>s have similar sizes, their tendencies are similar.
 <div id='img-2'></div>

@@ Line 252: / Line 252: @@
 ===5.3 Identification of the persistent topic===
-The next analysis concerned the identification of the persistent topics, which must exist over a given period. The persistent topic number is affected by σ<sub>1</sub>. The keyword of a topic always has a frequency of approximately 0.05, while a similarity of 0.1 means the topics have at least two keywords, and then it can be certain that they are in fact the same. Experiments have proven that an important turning point occurs at <math>\sigma_1=0.09</math>, corresponding to the 18 relatively persistent topics. The persistent topics have high accuracy and quality by manual validation.
+The next analysis concerned the identification of the persistent topics, which must exist over a given period. The persistent topic number is affected by <math>\sigma_1</math>. The keyword of a topic always has a frequency of approximately 0.05, while a similarity of 0.1 means the topics have at least two keywords, and then it can be certain that they are in fact the same. Experiments have proven that an important turning point occurs at <math>\sigma_1=0.09</math>, corresponding to the 18 relatively persistent topics. The persistent topics have high accuracy and quality by manual validation.
 There are 18 persistent topics with 4637 related posts. A total of 91281 users (28% of total users) were present in the following analysis, which greatly reduces the data size for further analysis. There are 257 related posts per persistent topic, and according to the minimum period (three months), they have only 86 posts per topic per month. This number is less than the size of the general topics retained in Section 5.2, which also reflects the persistent topics that do not have a high post rate, click rate or response rate and instead have their own characteristics of long duration.

@@ Line 238: / Line 238: @@
 ===5.2 Identification of the topics in specific periods===
-We used the LDA to identify the topic in specific months, setting <math display="inline">α=0.5,\, \beta =0.1</math>, topic number <math display="inline">Z=50</math> and Gionline Forum sampling iterations to 1000. Not all of each month's topics are related to the livelihood issues that this article focuses on, so these topics are omitted by the attribute filter described in Section 4.
+We used the LDA to identify the topic in specific months, setting <math display="inline">\alpha =0.5,\, \beta =0.1</math>, topic number <math display="inline">Z=50</math> and Gionline Forum sampling iterations to 1000. Not all of each month's topics are related to the livelihood issues that this article focuses on, so these topics are omitted by the attribute filter described in Section 4.
 After applying this attribute filter, there were a total of 978 topics with an average of 27 topics per month. The minimum number occurred in the 10<sup>th</sup> month with 9 topics and maximum was in the 6<sup>th</sup> month with 37 topics. To analyze the size of each topic, [[#img-1|Figure 1]] shows the statistics on the number of topics related posts. Setting <math>\sigma_{2}=0.05</math> retains more valid data for extracting persistent topics that is, if a title contains a keyword related to a certain topic, it will be retained. Eighty-two percent of retained topics ranged in size from 61 to 150 related posts.
@@ Line 249: / Line 249: @@
 | style="background:#efefef;text-align:left;padding:10px;font-size: 85%;"| '''Figure 1'''. The related posts number for each topic
 |}
 ===5.3 Identification of the persistent topic===

@@ Line 221: / Line 221: @@
 {| style="text-align: center; margin:auto;"
 |-
-| <math>C(x\rightarrow y)=\frac{A(x\rightarrow y)}{{\sum }_{j=1}^mA(x\rightarrow y_j)}</math>
+| <math>C(x\rightarrow y)=\frac{A(x\rightarrow y)}{{\displaystyle\sum }_{j=1}^mA(x\rightarrow y_j)}</math>
 |}
 | style="width: 5px;text-align: right;white-space: nowrap;" | (7)

@@ Line 30: / Line 30: @@
 Social network analysis (SNA) [3] can help us obtain the implicit characteristics of the users and information dissemination in a numerical manner. The forum topics are mainly divided into two categories: (1) emergency topics, which are characterized by a short duration with intense discussion; (2) persistent topics, characterized by long duration, typically closely related to one’s livelihood. Most studies have focused on the former, such as researching the discovery and prediction of online Forum hot topics and false information dissemination after emergencies [4]. There are two core issues that must be solved to identify key users in persistent livelihood topics: (1) extraction of persistent topics and (2) the identification of key users. To solve the first issue, we combine the time dimension and apply the latent Dirichlet allocation (LDA) topic model and the short text similarity assessment modelto discover the persistent topics [5]. To solve the second, SNA provides a series of node metrics (e.g., central, prestige, trust and connectivity). The node position assessment, proposed by Przemysław Kazienko, is a very effective method for analysis, but it is more suitable for the global network while ignoring the semantic factors. Therefore, we provided the sentiment weighted node position algorithm (SWNP) and applied it to the persistent topic network to sort the users’ influence.
-The algorithm must solve several problems. First, it must ensure that the extraction topic is related to the clustering results, so the algorithm uses the LDA model and the short text similarity assessment model for screening and gathering related posts while adopting adjacent time slice cross matching to ensure the topic sustainability on the timeline. After cataloging the posts, corresponding participants and replies relations, the persistent topic social network can be built and expressed as <math display="inline">(PTSN= V, E)</math>, where V and E represent the nodes and their relationships in the local network, respectively. It then identifies the critical nodes in the local network, which have the greatest amount of influence on the specific topic and other users. After attempting different methods on real three-year online Forum data, the SWNP is provided and compared to the typical method.
+The algorithm must solve several problems. First, it must ensure that the extraction topic is related to the clustering results, so the algorithm uses the LDA model and the short text similarity assessment model for screening and gathering related posts while adopting adjacent time slice cross matching to ensure the topic sustainability on the timeline. After cataloging the posts, corresponding participants and replies relations, the persistent topic social network can be built and expressed as <math display="inline">(PTSN= V, E)</math>, where <math display="inline">V</math> and <math display="inline">E</math> represent the nodes and their relationships in the local network, respectively. It then identifies the critical nodes in the local network, which have the greatest amount of influence on the specific topic and other users. After attempting different methods on real three-year online Forum data, the SWNP is provided and compared to the typical method.
 The rest of the paper is organized as follows. We briefly review related work in Section 2. We then present an overview of LDA and the short text similarity assessment model in Section 3. In Section 4, we propose persistent topic key person analysis in online Forum software, with detailed explanations. We discuss detailed experimental results on the research corpus in Section 5, and we conclude this paper in Section 6.

@@ Line 173: / Line 173: @@
 {| style="text-align: center; margin:auto;"
 |-
-| <math>sentiment_{i,j}=\frac{{\sum }_{k=1}^{n^j}O_{k,j}}{n^j}</math>
+| <math>sentiment_{i,j}=\frac{{\displaystyle\sum }_{k=1}^{n^j}O_{k,j}}{n^j}</math>
 |}
 | style="width: 5px;text-align: right;white-space: nowrap;" | (5)

@@ Line 155: / Line 155: @@
 | <math>\left\{ \begin{array}{c}
 POST_i={{\displaystyle\sum }_{t=j}^{j+s}POST}_i(ts)\\
-USER=_{}^i{{\displaystyle\cup }_{t=j}^{j+s}USER}_i(ts)
+USER=_{}^i{{\cup }_{t=j}^{j+s}USER}_i(ts)
 \end{array}\right.</math>
 |}