Kwok Yip Szeto

Learn More
Investment strategies as rules for buy and sell are introduced as conditional statements involving inequalities of various moving averages. Different conditional statements on moving averages are represented as strings, encodable as chromosomes in an approach based on genetic algorithm. The data mining of good investment strategies corresponds to the(More)
Universal topological properties of two-dimensional trivalent cellular patterns are found from shell analysis of soap froth and computer-generated Voronoi diagrams. We introduce a cluster analysis based on the shell model and find the universal relation ln(a/mu(2)) = A+Bln(mu(2)), with the generalized Aboav parameter a and second moment of the number of(More)
We report an experimental measurement of the temporal dependence of the area Aus in a twodimensional soap froth which has not been swept by the passage of soap films up to time t, as the froth coarsens from an initial time t0 within the scaling regime. We find Aus scales with the mean cell area kAl as Aus ~ kAl 0 , with a first-passage exponent u0 ­ 1.16 6(More)
The effect of random news on the performance of adaptive agents as investors in stock market is modelled by genetic algorithm and measured by their portfolio values. The agents are defined by the rules evolved from a simple genetic algorithm, based on the rate of correct prediction on past data. The effects of random news are incorporated via a model of(More)
The biological observation of the difference in the mutation rates of allele on different loci is implemented in genetic algorithm so that the mutation rate is both time and locus dependent. The performance of this new locus oriented adaptive genetic algorithm (LOAGA) is evaluated on the test problem of zero/one knapsack for various sizes. It is found that(More)
A new adaptive genetic algorithm using mutation matrix is introduced and implemented in a single computer using the quasi-parallel time sharing algorithm for the solution of the zero/one knapsack problem. The mutation matrix M(t) is constructed using the locus statistics and the fitness distribution in a population A(t) with N rows and L columns, where N is(More)