OPEN ACCESS
RÉSUMÉ. L’influence d’un blogueur sur Twitter véhicule une part importante de la pertinence de ses publications, elle peut donc être exploitée avantageusement en recherche d’information pour identifier les tweets pertinents. L’information sur l’influence des blogueurs est implicitement portée par la structure du réseau social de Twitter. Dans ce papier, nous exploitons cette structure pour mesurer l’influence. En particulier, nous proposons une approche adaptée de l’algorithme PageRank, pour l’analyse des liens du réseau social de rediffusion et le calcul de l’influence. Dans cette approche, l’influence du blogueur sur le réseau social est définie comme le rapport entre sa capacité à influencer d’autres blogueurs (influence imposée) et sa disposition à être influencé par d’autres blogueurs du réseau (influence subie). Nous évaluons notre score d’influence dans un contexte de RI sociale. Nous proposons alors un modèle de recherche d’information qui estime la pertinence d’un tweet comme une combinaison de sa pertinence thématique et de l’influence de son auteur. Ce modèle de recherche a été évalué sur la collection de tests TREC-Microblog 2011. Les résultats obtenus montrent tout son intérêt.
ABSTRACT. The blogger’s influence on Twitter conveys an important part of his publications relevance, it can therefore be advantageously exploited in information retrieval (IR), to identify relevant tweets. The information about blogger’s influence is implicitly driven by Twitter’s social network structure. In this paper, we exploit this structure to measure influence. In particular, we propose an adapted approach of the PageRank algorithm for analyzing retweet-based social network links and calculating influence. In this approach, the blogger’s influence in the social network is defined as the ratio of his ability to influence other bloggers (imposed influence) to his predisposition to be influenced by other bloggers in the network (suffered influence). We evaluate our ratio of influence in a context of social IR. We thus propose a social retrieval model which estimates the tweet relevance as a combination of its topic relevance and its author’s influence. This retrieval model has been evaluated on TREC-microblog 2011 test collection. The obtained results show all its interest.
MOTS-CLÉS : influence, réseau social de Twitter, recherche d’information sociale, PageRank.
KEYWORDS: influence, Twitter social network, social information retrieval, PageRank.
Anger I., Kittl C. (2011). Measuring Influence on Twitter. KNOW’11, Proceedings of the 11th International Conference on Knowledge Management and Knowledge Technologies. Article n° 31.
Azaza L., Kirgizov S., Savonnet M., Frame A. (2015). Evaluation de l’influence sur Twitter: Application au projet « Twitter aux Elections européennes 2014. Etudier le Web politique : regards croisés, May, Lyon.
Benjabeur L., Tamine L., Boughanem M. (2011). Un modèle de recherche d’information sociale dans les microblogs : cas de Twitter. Conférence sur les Modèles et l’Analyse des Réseaux: Approches Mathématiques et Informatique MARAMI’11.
Benjabeur L., Tamine L., Boughanem M. (2012). Active microbloggers: identifying influencers, leaders and discussers in microblogging networks. International Symposium on String Processing and Information Retrieval, p. 111-117.
Bogdan G., Zygmunt A. (2015). Finding Influential Bloggers. International Journal of Machine Learning and Computing. vol. 5, n° 2, April 2015.
Cano A. E., Mazumdar S., Ciravegna F. (2014). Social influence analysis in microblogging platforms – A topic-sensitive based approach. Semantic Web, vol. 5, n° 5, p. 357-372.
Cataldi M., Aufaure M.-A. (2014). The 10 Million follower fallacy: Audience size does not prove domain-influence on Twitter. Knowledge and Information Systems, p. 1-22.
Cha, M., Haddadi H., BenveutoF., Gummadi P. K. (2010). Measuring User Influence in Twitter: The Million Follower Fallacy. 4th International AAAI Conference on Weblogs and Social Media (ICWSM).
Ding Z., Jia Y., Zhou B., Han Y., He L., Zhang J. (2013). Measuring the spreadability of usersin microblogs. Journal of Zhejiang University - Science C, vol. 14, n° 9, p. 701-710.
Gayo-Avello D. (2013). Nepotistic relationships in Twitter and their impact on rank prestige algorithms. Information Processing and Management, vol. 49, n° 6, p. 1250-1280.
Herrera R., Cardenas J P., Alfaro R. (2014). User influence in on-line social networks. In WACS 2014, IV Workshop on Agents and Collaborative Systems. Jornadas Chilenas de Computacion(JCC). November 8-14, Talca, Chile.
Hiemstra D.(2001). Using Language Models for Information Retrieval. PhD thesis, Enschede. January.
Kleinberg J. (1999). Authoritative sources in a hyperlinked environment. Journal of the ACM (JACM), vol. 46, n° 5, p. 604-632.
Liu X., Shen H. Ma F., Liang W. (2014). Topical influential user analysis with relationship strength estimation in Twitter. In Zhi-Hua Zhou, Wei Wang, Ravi Kumar, Hannu Toivonen, Jian Pei, Joshua Zhexue Huang, Xindong Wu (Eds), 2014 IEEE International Conference on Data Mining Workshops, ICDM Workshops 2014, Shenzhen, China, December 14, p. 1012-1019.
Nargundkar A., Rao Y. S. (2016). InfluenceRank: A machine learning approach to measure influence of Twitter users. In International Conference on Recent Trends in information technology.
Page L., Brin S. Motwani R., Winograd T. (1999). The PageRank citation ranking:Bringing order to the web. Technical report, Stanford Digital Library.
Räbiger S., Spiliopoulou M. (2015). A framework for validating the merit of properties that predict the influence of a twitter user. Expert Syst. Appl., vol. 42, n° 5, p. 2824-2834.
Riquelme F., González-Cantergiani P. (2016). Measuring user influence on Twitter: A survey. In Journal of Information Processing and Management vol. 52, n° 5, p. 949-975, April.
Robertson S. E., Jones K. S. (1976). Relevance weighting of search terms. Journal of the American Society for Information Science, vol. 27, n° 3, p. 129-146, mai.
Romero D. M., Galuba W., Asur S., Huberman B. A. (2011). Influence and Passivity in Social Media. WWW’11, March 28-April 1, Hyderabad, India.
Sung J., Moon S., Lee JG. (2013). The influence in Twitter: Are they really influenced? In Longbing Cao, Hiroshi Motoda, Jaideep Srivastava, Ee-Peng Lim, Irwin King, PhilipS Yu, Wolfgang Nejdl, Guandong Xu, Gang Li, Ya Zhang (Eds), Behavior and Social Computing, vol. 8178 of Lecture Notes in Computer Science, p. 95-105. Springer International Publishing.
Tunkelang D. (2009). A Twitter Analog to PageRank, http://thenoisychannel.com/2009/ 01/13/a-twitter-analog-to-PageRank.
Weng J., Lim E.-P., Jiang J., He Q. (2010). TwitterRank: finding topic-sensitive influential Twitterers. Proceedings of the third ACM international conference on Web search and data mining, WSDM’10, New York, NY, USA, ACM, p. 261-270.
Zhang J., Zhang R., Su nJ., Zhang Y., Zhang C. (2016). TrueTop: A Sybil-Resilient System for User Influence Measurement on Twitter. In IEEE/ACM Transactions on Networking, vol. 24, n° 5, p. 2834-2846, October.
Zhang M., Sun C., Liu W. (2011). Identifying Influential Users Of Micro-Blogging Services: A Dynamic Action-Based Network Approach. PACIS 2011, Proceedings.223, http://aisel.aisnet.org/pacis2011/223.