Evaluation of Search Result of Document Search Based GA (DSEGA)

Since the 1940s the problem of information storage and retrieval has attracted increasing attention. Retrieve document becoming ever more difficult. With the advent of computers, a great deal of thought has been given to using them to provide rapid and intelligent retrieval systems. Although i has become easier to collect and store information in document collections, it has become increasingly difficult to retrieve relevant information from these large document collections. Genetic algorithms describe a set of optimization techniques that given a goal or fitness function are used to search a space for optimal points. The space in searched in a directed, stochastic manner, and the method of searching borrows some ideas from evolution. In practice, genetic algorithms have proven very effective in searching through complex, highly nonlinear, multidimensional search spaces. DSeGA system is an intelligent search agent toolkit at Faculty of Information Technology of Universiti Utara Malaysia. It is composed by a series of module that using information retrieval method and genetic algorithm. This toolkit does not tested by any standard test data collection. The aim of this research is to test DSeGA system with three standard data collection (Granfield, CACM and TIME). The finding of research is and evaluation of DSeGA system search result. It was discovered that DSeGA system cannot performed the way that the system should be. The conclusion of this research is DSeGA system need to be investigated to enhance the system performance.