1 X ^ ?2? i ??Vi ?MSE ?N i? K k2di^ where V

1 X ^ ?2? i ??Vi ?MSE ?N i? K k2di^ where V i ?is the predicted number of reMLN9708 price tweets of a tweet k posted by user i, N is the number of users, and Vi(k) is the real value. The predictions’ MSEs are presented in Table 3. Note that the DalfopristinMedChemExpress Dalfopristin settings for TwitterRank and TD-Rank are the same as in the other experiments.Table 3. Comparison of the predictions’ MSEs. PageRank MSE 29.95 LeaderRank 27.21 TwitterRank 18.03 TD-Rank 16.doi:10.1371/journal.pone.0158855.tPLOS ONE | DOI:10.1371/journal.pone.0158855 July 14,12 /Discover Influential LeadersAs shown in Table 3, PageRank and LeaderRank perform the worst, followed by TwitterRank. TD-Rank performs the best, outperforming the competition in terms of MSE, which indicates that the influential leaders discovered by our proposed algorithm are closer to the true situation.ConclusionIn this paper, we focused on discovering multi-topic influential leaders in social network. We proposed a multi-topic influence diffusion (MTID) model, which decomposes the influence of a particular user into two parts: direct influence, which is influence related to that user’s followers, and indirect influence, which is influence that is not restricted to direct followers. To cope with the definition of indirect influence, we introduced topic ground nodes that represent topic pools for establishing links between users. Moreover, to deal with the transition probability, we adopted a data-based approach that extracts the topic distribution from traces of tweets. Based on MTID, we further proposed a topic-dependent rank (TD-Rank) algorithm to identify the topic aware influential leaders. Finally, we conducted extensive experiments comparing the existing ranking algorithms using the Spark platform. The experimental results demonstrated that our proposed algorithm is more robust, more accurate and more sensitive to topic than previous algorithms. Our plans for future work include dealing with the dynamic structure of a following social network by incorporating a time factor into our model. We will also consider other influence measures in the future.Author ContributionsConceived and designed the experiments: XT QGM. Performed the experiments: XT SSY. Analyzed the data: XT. Contributed reagents/materials/analysis tools: YNQ. Wrote the paper: XT. Collected the data: SSY.
The aim of this study was to investigate the simultaneous presence of risk factors for noncommunicable diseases and the association of these risk factors with demographic and economic factors among adolescents from southern Brazil.OPEN ACCESS Citation: Nunes HEG, Gon lves ECdA, Vieira JAJ, Silva DAS (2016) Clustering of Risk Factors for NonCommunicable Diseases among Adolescents from Southern Brazil. PLoS ONE 11(7): e0159037. doi:10.1371/journal.pone.0159037 Editor: Andrea S. Wiley, Indiana University, UNITED STATES Received: August 27, 2015 Accepted: June 27, 2016 Published: July 19, 2016 Copyright: ?2016 Nunes et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: The authors have no support or funding to report. Competing Interests: The authors have declared that no competing interests exist.MethodsThe study included 916 students (14?9 years old) enrolled in the.1 X ^ ?2? i ??Vi ?MSE ?N i? K k2di^ where V i ?is the predicted number of retweets of a tweet k posted by user i, N is the number of users, and Vi(k) is the real value. The predictions’ MSEs are presented in Table 3. Note that the settings for TwitterRank and TD-Rank are the same as in the other experiments.Table 3. Comparison of the predictions’ MSEs. PageRank MSE 29.95 LeaderRank 27.21 TwitterRank 18.03 TD-Rank 16.doi:10.1371/journal.pone.0158855.tPLOS ONE | DOI:10.1371/journal.pone.0158855 July 14,12 /Discover Influential LeadersAs shown in Table 3, PageRank and LeaderRank perform the worst, followed by TwitterRank. TD-Rank performs the best, outperforming the competition in terms of MSE, which indicates that the influential leaders discovered by our proposed algorithm are closer to the true situation.ConclusionIn this paper, we focused on discovering multi-topic influential leaders in social network. We proposed a multi-topic influence diffusion (MTID) model, which decomposes the influence of a particular user into two parts: direct influence, which is influence related to that user’s followers, and indirect influence, which is influence that is not restricted to direct followers. To cope with the definition of indirect influence, we introduced topic ground nodes that represent topic pools for establishing links between users. Moreover, to deal with the transition probability, we adopted a data-based approach that extracts the topic distribution from traces of tweets. Based on MTID, we further proposed a topic-dependent rank (TD-Rank) algorithm to identify the topic aware influential leaders. Finally, we conducted extensive experiments comparing the existing ranking algorithms using the Spark platform. The experimental results demonstrated that our proposed algorithm is more robust, more accurate and more sensitive to topic than previous algorithms. Our plans for future work include dealing with the dynamic structure of a following social network by incorporating a time factor into our model. We will also consider other influence measures in the future.Author ContributionsConceived and designed the experiments: XT QGM. Performed the experiments: XT SSY. Analyzed the data: XT. Contributed reagents/materials/analysis tools: YNQ. Wrote the paper: XT. Collected the data: SSY.
The aim of this study was to investigate the simultaneous presence of risk factors for noncommunicable diseases and the association of these risk factors with demographic and economic factors among adolescents from southern Brazil.OPEN ACCESS Citation: Nunes HEG, Gon lves ECdA, Vieira JAJ, Silva DAS (2016) Clustering of Risk Factors for NonCommunicable Diseases among Adolescents from Southern Brazil. PLoS ONE 11(7): e0159037. doi:10.1371/journal.pone.0159037 Editor: Andrea S. Wiley, Indiana University, UNITED STATES Received: August 27, 2015 Accepted: June 27, 2016 Published: July 19, 2016 Copyright: ?2016 Nunes et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: The authors have no support or funding to report. Competing Interests: The authors have declared that no competing interests exist.MethodsThe study included 916 students (14?9 years old) enrolled in the.

Leave a Reply