المستخلص: |
In this paper, we explain the use of Genetic Algorithms to improve performance of Arabic information retrieval system, which based on vector space model. The performance was enhanced through the usage of an adaptive matching function, which obtained from a weighted combination of four similarity measures (Dot, Cosine, Jaccard and Dice), The genetic algorithms was used to optimize this matching functions. Optimization has been made through getting the best achievable combination of these weights. We used the precision as performance measure. The result of optimization process has shown a considerable improvement on the performance. The document collection was used is an Arabic documents collection.
|