Research Group Project and Description (cont'd) Review and Empirical Comparative Study of Parameterless Genetic Algorithms Nawwaf Kharma and Fadi Draidi In this paper we compare the performance of five Genetic Algorithms (GAs) that do not require the manual tuning of their parameters, and are thus called Parameterless GAs (pGAs). The five GAs selected for evaluation span the three most important categories of Parameterless GAs: Deterministic, Adaptive and Self-Adaptive. The five test functions used to evaluate the performance of the pGAs include multimodal and deceptive functions. We assess performance in terms of fitness, reliability, speed and memory load. The experimental results show that different Parameterless GAs are suited to searching different types of fitness surfaces. Surprisingly, the simplest Parameterless GA tested proved to be the best overall performer.         Fast and Robust GA-Based Ellipse Detection J. Yao and Nawwaf Kharma This study discusses a novel and effective technique for extracting multiple ellipses from an image, using a Genetic Algorithm with Multiple Populations (MPGA). MPGA evolves a number of subpopulations in parallel, each of which is clustered around an actual or perceived ellipse in the target image. The technique utilizes both evolution and clustering to direct the search for ellipses - full or partial. MPGA is explained in detail, and compared with both the widely used Randomized Hough Transform (RHT) and the Sharing Genetic Algorithm (SGA). In thorough and fair experimental tests, utilizing both synthetic and real-world images, MPGA exhibits solid advantages over RHT and SGA in terms of accuracy of recognition - even in the presence of noise or/and multiple imperfect ellipses in an image - and speed of computation.       SmallWorldz: a Client-Server Software System for Connecting People Tom Bailly-Salins and Nawwaf Kharma If we were to measure the diameter of the Earth by the time it takes an electronic message to circle the globe, then the world would appear as a very small place. However, today, it does not feel as such; and one reason it does not is the lack of effective human-centered communications' software meant to facilitate the creation of informal networks of friends or/and professionals (CommunityWear). In this paper, we review the most widely used types of Internet-based communications software, and then identify a set of elements that we believe are necessary for the success of human-centered software systems. Our work assumes that people want to connect with each other and understand the value of such connections; however, they require a convenient tool, which uses a "trust bridge" to facilitate the establishment of such connections. We describe our own (copyrighted) attempt at creating such a system, dubbed "SmallWorldz". The architecture and operation of SmallWorldz are described, together with a built-in evaluation tool, which allows (super-) users to measure the effectiveness of the software.     Page 1, Page 2, Page 3, Page 4   DEC Lab Home Last Updated April 6, 2004