CLASS-BASED DISTRIBUTED EVOLUTIONARY ALGORITHM FOR ASSET MANAGEMENT AND TRADING

摘要

The server computer and the plurality of client computers form a network computing system configured to continue to evaluate the performance characteristics of multiple genes generated using scalable and software applications. Each client computer continues to periodically receive data associated with the genes stored in memory. Using this data, the client computer evaluates the performance characteristics of the gene by comparing the solution provided by the gene with the periodically received data associated with that gene. Thus, the performance characteristics of each gene may be updated and changed by each periodically received data. The performance characteristics of genes define conformity. The gene may be a virtual property trader who has recommended a trading option. A gene may initially be assigned to a different class to improve convergence, but later it may be determined to merge with the genes of other classes to improve diversity.