Evaluating class association rules using Genetic Relation Programming

A portfolio selection model using genetic relation algorithm and genetic network programming

“…Concretely speaking, the efficiency of GNP for generating stock trading rules has been confirmed in our previous studies [32][33][34][35]. Also, GRA is originally developed to reduce a large class association rule set for data mining [31]. Unlike strings for solution representation in GA and trees in GP, GRA has the ability to express complex events compactly with graph structures, evading black-box issues and exhaustive mathematical properties needed for encoding in other natural inspired algorithms.…”

“…In addition, the buried noise and complex dimensionality of the stock market data make it difficult to learn or re-estimate the ANN parameters [28]. Due to such kinds of bottlenecks, GNP [29,30] and GRA [31] are proposed to solve these problems. Concretely speaking, the efficiency of GNP for generating stock trading rules has been confirmed in our previous studies [32][33][34][35].…”

Diversifying risk in portfolios using a variable‐size genetic relation algorithm

“…In general terms, the aim of vs‐GRA is to obtain a compact set of diverse events from an observed environment in a dynamic manner [15]. In the portfolio diversification context, an event refers to an asset, and environments refer to financial markets.…”

“…2, which expresses a portfolio of assets including diverse asset classes such as stocks, bonds, and currencies in the portfolio diversification context. Different graph configurations are also possible for other problem domains [15]. The genotype structure encodes the information contained in the individual, whose internal data structure is a string, just as in any other evolutionary computing technique.…”