Optimization of knowledge discovery methods for library data mining based on augmented learning

Among the various services of smart libraries, knowledge discovery services are becoming more and more popular among users.Knowledge discovery based on data mining supports users to obtain convenient resources, as a result, this paper optimizes the library knowledge discovery method, utilizes the DDGP reinforcement learning algorithm for enhancement learning, and proposes a dual experience pool structure and hierarchical experience playback mechanism to improve the DDPG algorithm, and realizes the library accurate recommendation model based on enhancement learning.Experiments are carried out on different datasets, and the results of each evaluation index of the improved DDPG algorithm in this paper are better than those of the comparison methods, with the hit rate and the cumulative gain of the normalized discount improved by 0.132~0.380and 0.074~0.308compared with that of the DDPG algorithm, and the superior performance of the accuracy, the recall and the F-value are also obtained under different recommendation numbers.Experiments show that the library precision recommendation method based on augmented learning in this paper has better resource recommendation accuracy and can provide users with more personalized knowledge discovery services.