Modeling real-world phenomena is a focus of many science and engineering efforts, from ecological modeling to financial forecasting. Building an accurate model for complex and dynamic systems improves understanding of underlying processes and leads to resource efficiency. Knowledge-driven modeling builds a model based on human expertise, yet is often suboptimal. At the opposite extreme, data-driven modeling learns a model directly from data, requiring extensive data and potentially generating overfitting. We focus on an intermediate approach, model revision, in which prior knowledge and data are combined to achieve the best of both worlds. We propose a genetic model revision framework based on tree-adjoining grammar (TAG) guided genetic programming (GP), using the TAG formalism and GP operators in an effective mechanism making data-driven revisions while incorporating prior knowledge. Our framework is designed to address the high computational cost of evolutionary modeling of complex systems. Via a case study on the challenging problem of river water quality modeling, we show that the framework efficiently learns an interpretable model, with higher modeling accuracy than existing methods.

Water quality is an important global issue, requiring effective management, which needs good predictive tools. While good methods for lake water quality prediction have previously been developed, accurate prediction of river water quality has hitherto been difficult. This project combines process-model and data mining approaches through evolutionary methods, resulting in tools for more effective water management. Although the work is still in its preliminary stages, error rates of the predictive models are already around half those resulting from representative applications of either pure process-based or pure data mining approaches.

In this paper, we investigate the value of a new selection mechanism that is inspired from biochemistry, namely, vesicular selection. We test its effectiveness when used in evolutionary algorithms on a number of benchmark problems in both static and dynamic environments. The results suggest beneficial attributes for the new selection mechanism.

In this paper, we give a framework to reduce delivery failure influence in overlay multicast. Unlike previous efforts based on duplicated forwarding, our scheme builds another type of overlay data delivery tree (DDT), which is adaptively reconstructed based on the number of measured packet delivery ratio while guaranteeing end-to-end delay bound. Through practical simulation results, we analyzed packet delivery ratio, control overhead, end-to-end delay and number of replicated packets.

Various ALM (application layer multicast) mechanisms have been suggested to remove the barriers that block deployment of IP multicast. However, most ALM mechanisms do not consider the host group model, which is not a problematic portion, but advantageous feature of traditional IP multicast in terms of scalability, robustness, security, and transparency. Furthermore, those mechanisms build an overlay multicast network for data delivery mostly based on a simple measurement, with a few metrics, without directly considering the IP network topology. Thus, they might suffer some inefficiency, such as long join delay and large control overhead. We propose a scheme to take advantages of both IP multicast and ALM, based on two principal design factors. By adopting the host group model, our scheme can inherently acquire network topology information used to get rid of application level routing inefficiency, and to gain additional efficiency for many group members. Also, we exploit topology-awareness to reduce join delay and localize group members so that controls for setting up and maintaining overlay networks can be limited in a domain. Overall network performance is enhanced with our scheme by building an overlay network based on performance factors associated with network characteristics using a hierarchical structure.