An intelligent information retrieval algorithm based on knowledge discovery and self-organizing feature map neural network

Abstract

Information retrieval is usually referring to the text information retrieval, including information storage, organization, performance, many aspects, such as query, access and its core is the text indexing and retrieval of information. Under the trend of intelligent data analysis and mining, in this paper, we propose a novel information retrieval algorithm based on knowledge discovery and self-organizing feature map neural network. Knowledge discovery is one of the major intellectual activities of human, the current knowledge discovery activities are increasingly based on network data resources and environment. Enhance semantic correlation method, that is, on the basis of the existing association, found the correlation between the data source, a new connection between different sources of data, or further connection, this process is the process of knowledge discovery activities. For enhancement, we introduce the self-organizing feature map neural network into the method to integrate the semantic information. Since Kohonen self-organizing neural network is put forward, the self-organizing feature map algorithm as a kind of very effective clustering method, in the vector quantization and pattern recognition has been widely research and application. With the reasonable of the mentioned techniques, we propose the enhanced retrieval algorithm. The experimental simulation proves that our method obtains higher robustness and accuracy compared with the other state-of-the-art algorithms.