Research on the rapid discovery of information resources in the training and management of big data construction

It is the key content of the current military information system construction to quickly and accurately provide users with the required information resources from the massive military training big data. With the rapid increase of training data and the rapid expansion of the network scale, the centralized data storage and search in the past is likely to form a performance bottleneck. To this end, this paper proposes an information resource discovery method based on adaptive k-nearest neighbor clustering. This method groups data resources with similar characteristics into the same category and provides an efficient inter-class message forwarding mechanism, which can reduce the search scale of the resource discovery and the diffusion scope of resource information update, so as to improve the performance of resource discovery. The simulation results show that compared with other existing methods, the information resource discovery method proposed in this paper can not only discover resources quickly and accurately, but also adapt to the development of network scale, so it has higher resource discovery efficiency and better adaptability.