Download as PDF, PPTX
Uploaded byLeonid Zhukov
Information cascades
1) The document discusses social networks and how information spreads through cascades and can enable viral marketing. 2) It describes how information cascades occur when people make decisions based on the actions of others rather than their own information. 3) Diffusion models are discussed as a way to model how information or influence spreads from person to person through social networks based on their connectivity and thresholds for adopting new information.
In this document
Powered by AI
Introduction to the concepts of social networks, information cascades, and viral marketing.
Exploring the decision-making process in social contexts, particularly conformity in restaurant choices.
Definition of information cascades where individuals mimic others’ choices regardless of their private signals.
Discusses conditions of decision-making in information cascades, including incomplete information and rationale.
How information spreads through network structures using distribution mechanisms as the medium.
Characteristics of social networks such as power law distribution and small-world phenomena.
Presentation of a diffusion model simulating how information spreads like a virus through networks.
Sequential steps in the diffusion model process, emphasizing connection patterns and coverage timing.
Introduction to the threshold model and its applications in opinions, purchasing decisions, and innovations.
Explains the threshold concept with a simple example illustrating decision-making behavior.
Stepwise explanation of the threshold model highlighting decision dynamics, incomplete cascades, and network effects.
Importance of strategic source locations in enhancing information spread and competitive advantage.
Exploration of dynamics in viral marketing specifically related to product recommendation networks.
Analysis of social media interactions like retweets and mentions as part of information diffusion.
Conclusion of the presentation without specific content.
More Related Content
Information cascades
- 1. Social Networks Information Cascades and Viral Marketing Saturday, September 8, 12
- 2. Which restaurant to choose? conformity Saturday, September 8, 12
- 3. Information cascade making the same decision that the others have made, independently of the own private information signals Saturday, September 8, 12
- 4. Information cascade Sequential decision making Incomplete information The choice of previous participants is known (but no the reasons!) Limited selection space Rational decision making A posteriori probability (Bayes’ theorem) Saturday, September 8, 12
- 5. Information spread Medium: network structure Model: distribution mechanism Saturday, September 8, 12
- 6. Universal structure of social networks Power low degree distribution Gigantic connected component Small world phenomenon Large clustering coefficient Hierarchical structure Saturday, September 8, 12
- 7. Diffusion model method - diffusion virus model “infected” on contact probability depends on immunity can model news gossips Saturday, September 8, 12
- 8. Diffusion model Step 1 Saturday, September 8, 12
- 9. Diffusion model Step 2 Saturday, September 8, 12
- 10. Diffusion model Step 3 Saturday, September 8, 12
- 11. Diffusion model Step 4 Saturday, September 8, 12
- 12. Diffusion model Step 5 Complete coverage Process time depends on the source Based on connectivity pattern Saturday, September 8, 12
- 13. Threshold model neighbors “opinion” decision threshold information cascade can model beliefs propagation A,B - types of behavior purchasing decisions q –decision threshold spread of innovations If fraction of neighbors with A greater than q, accpet A Saturday, September 8, 12
- 14. Threshold model threshold= 1/2 Saturday, September 8, 12
- 15. Threshold model Step 1 2 sources threshold= 2/5 Saturday, September 8, 12
- 16. Threshold model Step 2 Saturday, September 8, 12
- 17. Threshold model Step 3 Saturday, September 8, 12
- 18. Threshold model Step 4 incomplete cascade depends on network topology strongly depends on source selection Saturday, September 8, 12
- 19. Complete cascades Strategic source placement Well connected Inside various communities Increase of competitive advantage reduce threshold level Saturday, September 8, 12
- 20. “The Dynamics ofViral Product recommendation network Marketing” J. Leskovec et al, 2007 (DVDs) Saturday, September 8, 12
- 21. #newsjp Twitter retweet #ocra mention “Truthy” Project. Center for Complex Networks and System Research. Indiana University. #iraq Saturday, September 8, 12
- 22.