Influence

Propagated objects

• Topic effect
  • Jie Tang, Jimeng Sun, Chi Wang, and Zi Yang. Social Influence Analysis in Large-scale Networks. In SIGKDD'09. pp. 807-816.
    Propose a factor graph model to learn topic influence and apply it into expert finding. Data: coauthor and citation network (arnetmner.org) and Wikipedia (wikipedia.org).

  • Lu Liu, Jie Tang, Jiawei Han, Meng Jiang and Shiqiang Yang. Mining topic-level influence in heterogeneous networks. In CIKM'10. pp.199-208.
    Propose a topic model to learn topic influence. Data: Twitter (twitter.com), Digg(digg.com) and citation network (www.aminer.cn).

  • Jainshu Weng, Ee-Peng Lim, Jing Jiang and Qi He.TwitterRank: Finding Topic-sensitive Influential Twitterers. In WSDM'10.
    Propose an extensive PageRank to learn topic level influence score for each user. Data: Twitter (twitter.com)

  • Nicola Barbieri, Francesco Bonchi and Giuseppe Manco. Topic-Aware Social Influence Propagation Models. In ICDM'12. pp.81-90.
    Propose a topic extensive IC and LT model to learn topic influence and apply it into influence maximization. Data: Digg (digg.com) and Flixster (flixster.com)
  • Cindy Xide Lin, Qiaozhu Mei, Jiawei Han, Yunliang Jiang, and Marina Danilevsky. The Joint Inference of Topic Diffusion and Evolution in Social Communities. In ICDM'11.
    Propose a mixture model to co-infer the topic evolution and the the influence strength between nodes. Data: DBLP and Twiiter
• Sentiment effect
  • Wei Chen, Alex Collins, Rachel Cummings, Te Ke, Zhenming Liu, David Rincon, Xiaorui Sun, Yajun Wang, Wei Wei, and Yifei Yuan. Influence Maximization in Social Networks When Negative Opinions May Emerge and Propagate. In SDM 2011.
    Propose a IC model with negative opinions and apply it into influence maximization. Data: Academic nework (arXiv.org) and Wikipedia (wikipedia.org)
  • Daifeng Li, Xin Shuai, Guozheng Sun, Jie Tang, Ying Ding, and Zhipeng Luo. Mining Topic-Level Opinion Influence in Microblog. In CIKM'12. pp. 1562-1566.
    Propose a topic model to learn the positive/negative influence on different topics.
• Effect from other propagated objects
  • Seth A. Myers and Jure Leskovec. Clash of the Contagions: Cooperation and Competition in Information Diffusion. In ICDM 2012.
    Propose a diffusion model to determine the probability of a user adopting an object based on what other objects the user was previously exposed to. Data: Twitter (twitter.com)

• Time
  • Time affects the mechanim of influence
    • Amit Goyal, Francesco Bonchi and Laks V. S. Lakshmanan. Learning Influence Probabilities In Social Networks. In WSDM 2010.
      Propose Discret Time Model and Continuous Time Model to learn influence. Data: Flickr (flickr.com)
    • Kazumi Saito, Masahiro Kimura, Kouzou Ohara and Hiroshi Motoda. Learning Asynchronous-Time Information Diffusion Models and its Application to Behavioral Data Analysis over Social Networks. CoRR abs/1204.4528 (2012).
      Add time-delay and link-delay into IC model and LT model. Data: Wikipedia (wikipedia.org)
    • Emmanuel Maruani, Michel Grabisch and Agnieszka Rusinowska. A study of the dynamic of influence through differential equations. RAIRO - Operations Research. 2012. pp.83-106.
      Study the dynamic of some ifnluence functions (e.g., the majority and guru influence functions) introducted orginally in the discrete model by using differential euqations.
  • Influence strength between individual changes over time
    • Manuel Gomez-Rodriguez, Jure Leskovec and Bernhard Schölkopf. Structure and Dynamics of Information
      Pathways in Online Media. In WSDM' 13.
      Infer influence strength between edges over at different time. Data: blog post network.
    • Wei Pan, Manuel Cebrián, Wen Dong, Taemie Kim and Alex Pentland. Modeling Dynamical Influence in Human Interaction Patterns.CoRR.2010.
      Propose a HMM-like model to learn influence strength between individuals at different time and apply it to predict turn taking in the a group discussion (Here influence indicates how people influence each others' turn taking). Data:Audio recording for group discussion.
• Network
  • Network structure effect
    • Glen Jeh and Jennifer Widom, SimRank: a measure of structural-context similarity. In KDD, pp 538-543, 2002.
      Measure the similarity of two nodes based on their neighbors' structures.
  • Signed network effect
    • Yanhua Li, Wei Chen, Yajun Wang and Zhi-Li Zhang. Influence Diffusion Dynamics and Influence Maximization in Social Networks with Friend and Foe Relationships. In WDM 2013.
      Propose a voter model in signed network with postive and negative relationships and apply it into influence maximization. Data: Epinions (Epinions.com) and Slashdot (slashdot.org)
  • Indirect relationship effect
    • Xin Shuai,Ying Ding,Jerome Busemeyer, Shanshan Chen, Yuyin Sun and Jie Tang. Modeling Indirect Influence on Twitter. In IJSWIS 2013.
      They observe that the overall trend of reweeting probability increases with the number of spreaders, however, decreases locally. They use quantum cognition theory to interpret the local anomaly.
• External effect
  • Seth Myers, Chenguang Zhu and Jure Leskovec. Information Diffusion and External Influence in Networks. In KDD 2012.
    They model information diffusing to a node via the links of the social network or throgh the influence of the external source. Data: twitter (twitter.com)

• Influence Maximization
  • D. Kempe, J. Kleinberg, and E. Tardos. Maximizing the spread of influence through a social network. In KDD’03, pp. 137–146, 2003.
    General greedy algorithm
  • Leskovec J, Krause A, Guestrin C, Faloutsos C, VanBriesen J, Glance NS (2007) Cost-effective outbreak detection in networks. In KDD '07.
    Lazy forward optimization.
  • W. Chen, Y. Wang, and S. Yang. Efficient influence maximization in social networks. In KDD'09, pp. 199-207, 2009.
    Degree discount Heuristic.
• Identify influential nodes
  • Kang Zhao, Greta Greer, Baojun Qiu, Prasenjit Mitra, Kenneth Portier, and John Yen. Finding influential users of an online health community: a new metric based on sentiment influence. CoRR. 2012.
    Observe that after at least one responding reply is received, 75% thread originators who started with negative sentiment expressed positive subsequent sentiment; 85% thread originators who started with positive sentiment stayed postive. They find influential users according to the influential responding replies. Data: online cancer survivors network (http://csn.cancer.org)
• Infer social tie
  • Seth A. Myers and Jure Leskovec. On the Convexity of Latent Social Network Inference. In NIPS 2010.
    Manuel Gomez-Rodriguez, Jure Leskovec and Andreas Krause. Inferring Networks of Diffusion and Influence. IN KDD 2010.
    Define a maximum likelihood function according to the some cascade models. Data: email communication network, scientists collaboration network, person-person recommendation network and blog post network
• Community detection
  • Angel Stanoev and Daniel Smilkov and Ljupco Kocarev.
    Community detection is performed by finding all natural leaders and all nodes on which they influence. CoRR, 2011. Do community detetion through finding all natural leaders and all nodes on which they influence. Data: Zachary karate network (W. W. Zachary. An information flow model for conflict and ssion in small groups, Journal of Anthropological Research 33, 452-473, 1977).
• Predict user behaviors
  • Eunjoon Cho, Seth A. Myers and Jure Leskovec. Friendship and Mobility: User Movement In Location-Based Social Networks. KDD 2011.
    Observe that the influence of friendship on one’s mobility is two times stronger than the influence of mobility on creating new friendships. They develop a model to predict future movements of an individual by incorporating temorally and geographically periodic movement with the social network structure. Data: checkin data from Gowalla and Brightkite, a dataset of cell phone location trace data.
  • Stephen Guo, Mengqiu Wang, Jure Leskovec.The Role of Social Networks in Online Shopping: Information Passing, Price of Trust, and Consumer Choice. ACM Conference on Electronic Commerce (EC), 2011.
    Observe that the more mutual contacts a pair of users has, the greater the likelihood that they engaged in a commercial transaction. They predict which seller a buyer will choose to buy an item from by using SVM-rank. Data: taobao (www.taobao.com).
  • Jure Leskovec, Lada Adamic and Bernardo Huberman. The Dynamics of Viral Marketing. ACM Conference on Electronic Commerce (EC), 2006.
    Observe that the probablity of purchasing a product increases with the number of recommendations received, but quickly saturates to a constant. They predict users' phurchase behaviors and find that personal recommendations are most effective in small, densely connected communities enjoying expensive products.