A machine learning approach to document retrieval: an overview and an experiment

We provide an overview of artificial intelligence techniques and then present a machine learning based document retrieval system we developed. GANNET (Genetic Algorithms and Neural Nets System) performed concept (keyword) optimization for user-selected documents during document retrieval using genetic algorithms. It then used the optimized concepts to perform concept exploration in a large network of related concepts through the Hopfield net parallel relaxation procedure. Our preliminary experiment showed that GANNET helped improve search recall by identifying the underlying concepts (keywords) which best describe the user-selected documents.<
>