This web-site offers my open source implementation of the Community Optimization (CO) algorithm.

The CO algorithm draws its inspiration from the collective intelligence emerged by web communities and implements a behavioral model derived from the human behavior that can be observed within certain types of these web communities like, e.g., the Wikipedia authors. Wikipedia's article base is generalized as a "knowledge-base". This knowledge-base represents the problem to be solved and is realized as a real-valued vector. The vector components represent the different topics contained in this knowledge-base.

CO employs a community of persons to optimize the different topics of the knowledge-base. For each topic, each person has his own knowledge and expertise. While optimizing, a person contributes to a given topic based on his expertise (Contribution Rule). The expertise increases with the success of his contributions. Furthermore, all persons learn from the knowledge-base as well to improve their own knowledge (Learning Rule). An iteration of CO performs both rules in a sequence.

For a more detailed description of CO, the reader is referred to my introductory paper.

My implementation of CO can be downloaded as open source module written in ANSI-C under the terms of GNU GPL. It successfully compiles on

• MS Visual VC8    (MS Windows)
• Borland C++ Builder 5    (MS Windows)
• cc    (Linux, Solaris)
• g++    (Linux, Solaris)

Other compilers (platforms) were not tested, since they are out of my reach. The university does not have many commercial compilers. But it compiles with cc and g++. Thus, it should be usable with most compilers out there.

Download from SourceForge (.zip)

Note that there is only one set of files, which was developed under Windows using Windows newlines. On some platforms it could be necessary to convert these newlines, if one wants to develop based on my files. But most editors offer this functionality and some even work with both in parallel.

Abstract — In recent years a number of web-technology supported communities of humans have been developed. Such a web community is able to let emerge a collective intelligence with a higher performance in solving problems than the single members of the community. Based on the successes of collective intelligence systems like Wikipedia, the web encyclopedia, the question arises, whether such a collaborative web community could also be capable of function optimization. This paper introduces an optimization algorithm called Community Optimization (CO), which optimizes a function by simulating a collaborative web community, which edits or improves an article-base, or, more general, a knowledge-base. In order to realize this, CO implements a behavioral model derived from the human behavior that can be observed within certain types of web communities (e.g., Wikipedia or open source communities). The introduced CO method is applied to four well-known benchmark problems. CO significantly outperformed the Fully Informed Particle Swarm Optimization as well as two Differential Evolution approaches in all four cases especially in higher dimensions.