This past week brought two publication deadlines, a conference submission deadline, and preparation for a software demo at Harvard. Needless to say, I am exhausted, but it was well worth the effort.
The first publication is a 2-page summary of work I’ve been doing with Prof. Larry Yaeger looking at speciation mechanisms in artificial life simulations. This was a condesnation of a paper submission for the Genetic and Evolutionary Computing Conference, and I’m really pleased with how much we were able to squeeze in. Abstract, citation, and link follow:
Artificial life simulations can yield distinct populations of agents representing different adaptations to a common environment or specialized adaptations to different environments. Here we apply a standard clustering algorithm to the genomes of such agents to discover and characterize these subpopulations. As evolution proceeds new subpopulations are produced, which show up as new clusters. Cluster centroids allow us to characterize these different subpopulations and identify their distinct adaptation mechanisms. We suggest these subpopulations may reasonably be thought of as species, even if the simulation software allows interbreeding between members of the different subpopulations. Our results indicate both sympatric and allopatric speciation are present in the Polyworld artificial life system. Our analysis suggests that intra- and inter-cluster fecundity differences may be sufficient to foster sympatric speciation in artificial and biological ecosystems.
Jaimie Murdock and Larry Yaeger. Genetic Clustering for Species Identification. In Proceedings of the Genetic and Ecolutionary Computation Conference (GECCO) 2011. Dublin, Ireland, 2011. [paper]
The second publication is an expansion of the work on ontology evaluation presented last year at the 2010 International Conference on Knowledge Engineering and Ontology Development (KEOD) in Valencia, Spain. We’ve completely rewritten the section on our volatility score, and tightened up the language throughout. The 20-page behemoth will be published as a chapter in an upcoming volume of Springer-Verlag’s Communications in Computer and Information Science (CCIS) series. Abstract, citation, and link follow:
Ontology evaluation poses a number of difficult challenges requiring different evaluation methodologies, particularly for a "dynamic ontology" generated by a combination of automatic and semi-automatic methods. We review evaluation methods that focus solely on syntactic (formal) correctness, on the preservation of semantic structure, or on pragmatic utility. We propose two novel methods for dynamic ontology evaluation and describe the use of these methods for evaluating the different taxonomic representations that are generated at different times or with different amounts of expert feedback. These methods are then applied to the Indiana Philosophy Ontology (InPhO), and used to guide the ontology enrichment process.
Jaimie Murdock, Cameron Buckner and Colin Allen. Evaluating Dynamic Ontologies. Communications in Computer and Information Science (Lecture Notes). Spencer-Verlag. 2011. [chapter]