![]() and Rust, Roland T., Agent-Based Modeling in Marketing: Guidelines for Rigor (June 10, 2011). Stonedahl, Forrest, William Rand, and Uri Wilensky (2010), "Evolving Viral Marketing Strategies," Genetic and Evolutionary Computation Conference (GECCO), July 7-11, Portland, OR, USA Ideally it would be possible to set a number of infected turtles, and choose proportions of centrality measures to infect these turtles. These could be extended as well.įinally, the model only allows users to choose one centrality measure by which to initially infect turtles. ![]() If the model were to fully implement the NPV of a model, the value of each infected node would depend on at what time it was infected.įurther, the model currently allows for only two different network types and four centrality measures. Currently the model simply measures how fast a virus spreads on a network, but does not take into account how fast it is spread to each individual node which is one way to calculate the net present value (NPV) of a strategy. The model was inspired by the Local Viral Marketing Problem (Stonedahl, Rand & Wilensky 2010) of an adoption network. Are there particular measures that might result in a faster spread in particular types of networks, but not in others? Why do you think that is? EXTENDING THE MODEL Try to set up different kinds of networks, and select the initially seeded turtles by different centrality measures. When you click SETUP, the network will be generated, and when you click GO, nodes will start infecting their neighbors. Then choose how many turtles are initially seeded and by what centrality measure these turtles are selected. The model stops when the whole market is saturated. At each tick, each turtle has a small chance of adopting the product on their own (1%), and a larger chance of adopting the product if any of their friends have adopted the product (0.5 * the ratio of neighbors who have adopted to the number of all neighbors). A number of turtles are seeded with the product at time 0. When the model is set up, 500 turtles are created and connected in a network of the type selected (preferential attachment or a random network). Its purpose is to allow you to explore the relationship between different centrality measures, and different network types, and see if the interactions between them make for faster or slower spread of the product. This model is a simple experiment that allows researchers to examine how to best seed a network to maximize the adoption rate of a product by using viral marketing. The model, as well as any updates to the model, can also be found on the textbook website. This model is in the IABM Textbook folder of the NetLogo Models Library. Introduction to Agent-Based Modeling: Modeling Natural, Social and Engineered Complex Systems with NetLogo. This model is from Chapter Eight of the book "Introduction to Agent-Based Modeling: Modeling Natural, Social and Engineered Complex Systems with NetLogo", by Uri Wilensky & William Rand. Note: If you download the NetLogo application, every model in the Models Library is included. (back to the library) Simple Viral Marketing NetLogo Models Library: Simple Viral Marketingīeginners Interactive NetLogo Dictionary (BIND)
0 Comments
Leave a Reply. |