Finally, it is also possible to code agent-based models using general-purpose programming languages directly. There is also a Wikipedia page set up by Nikolai and Madey (2009) which provides an up-to-date comparison of agent-based software toolkits. Another recent review that assesses and compares NetLogo with other platforms has been published by Kravari and Bassiliades (2015). (2017), who compare 85 tools using a convenient tabular and chart format, and deem NetLogo both easy to use and also appropriate to execute medium/large-scale simulations. To our knowledge, the most up-to-date and comprehensive review of agent-based simulation software has been conducted by Abar et al.xxii) as lead developer for over a decade. It is also important to acknowledge Seth Tisue, who " worked meticulously to guarantee the quality of the NetLogo software" (Wilensky and Rand, 2015, p. NetLogo was created by Uri Wilensky and is under continuous development at the Northwestern's Center for Connected Learning and Computer-Based Modeling.The Mathematica link comes bundled as part of the latest NetLogo releases.Ĭonversely, one can also call R, Python and Matlab commands from within NetLogo using the R-Extension (Thiele and Grimm, 2010), the NetLogo Python extension (Head, 2018) and MatNet (Biggs and Papin, 2013) respectively. The connector PyNetLogo (Jaxa-Rozen and Kwakkel, 2018) provides the same functionalty for Python, and the so-called Mathematica link (Bakshy and Wilensky, 2007) for Mathematica. ( 2012a, 2012b, 2014)), it is possible to run and control NetLogo models from R, execute NetLogo commands, and obtain any information from a NetLogo model. Specifically, using an R package called RNetLogo ( Thiele (2014) Thiele et al. NetLogo can be linked with advanced software tools like R (R Core Team, 2019), Python (Python Software Foundation, 2019), Mathematica (Wolfram Research, Inc., 2019) or Matlab (The MathWorks, Inc., 2019). NetLogo is now a powerful tool widely used in science and we recommend it strongly, especially for those new to modeling and programming but also for serious scientists with software experience. Java or Objective-C, and can often reduce programming efforts significantly when compared with other languages. NetLogo language is definitely simpler to use than e.g. NetLogo also has the big advantage over pseudo-code of being executable, so the user can run and test the examples. Since NetLogo was designed to be easily readable, we believe that NetLogo code is about as easy to read as any pseudo-code we would have used. As a matter of fact, NetLogo language could perfectly be used as pseudo-code to communicate algorithms implemented in other languages. One characteristic that makes the NetLogo language easy to learn is that it is remarkably close to natural language. Someone with programming experience could reduce the estimated time to 1-2 days. To be concrete, we would estimate that an average scholar without previous coding experience can learn the basics of the language and be in a position to write a simple agent-based model after 2-4 days of work. All reviews of the software highlight how easy it is to learn. The language used to code models within NetLogo –which is also called NetLogo– has been designed following a “Low Threshold, No Ceiling” philosophy (Wilensky and Rand, 2015). NetLogo stands out as the quickest to learn and the easiest to use. Useful to conduct experiments with real people and for participatory modeling.Extensions to fulfill specialised needs.Great support and active user community.Multiplatform and online execution of models.Automatic exploration of parameter space.Possibility to interact with the model at runtime.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |