Association retrieval and recommendation of agricultural environmental sensing data based on knowledge map

Abstract

When the current general methods are used to retrieve agricultural environmental sensing data, there is no knowledge map constructed, and there are often problems sunch as low recall rate, low precision rate and low retrieval efficiency. In order to retrieve agricultural environmental sensing data efficiently, accurately and comprehensively, the method of association retrieval and recommendation of agricultural environmental sensing data based on knowledge map was proposed. The NLP algorithm was used to associate knowledge concepts of various agricultural environmental sensing data to construct the knowledge map of agricultural environmental sensing data. The knowledge map was pruned and expanded to reduce the amount of related calculations, and improved the retrieval efficiency of agricultural environmental sensing data. On the basis of the knowledge map of agricultural environmental sensing data, the vector space retrieval model was further used to realize the association retrieval and recommendation of agricultural environmental sensing data. To compared with the data retrieval method based on the quantification of the product within the cluster and the agricultural environment before the optimization of the knowledge map Sensing data association retrieval and recommendation method, the relevant experimental results showed that when the number of retrieved data sets increased, the recall rate, precision rate, and retrieval efficiency of the agricultural environmental sensing data association retrieval and recommendation method based on the knowledge map optimized after the knowledge map constructed in this paper were higher than the other two retrieval methods, which had high practical applicability and development prospects.