Answers are grounded in this paper and cite their sources.
The scale of electronic information systems continues to expand, and the management complexity increases. Deep learning provides a new approach for intelligent management of the system. This study constructs the DL-IMM model, optimizes Vanilla RNN using ECS, and conducts anomaly detection on system operation data. The data is divided into training set, validation set and test set according to the time sequence. PCA retains 4 principal components, which cumulatively explain 96.7% of the variance. The accuracy of the test set is 99.14%, the precision is 98.9%, the recall rate is 99.11%, and the F1-score is 98.97%. The validity of the model is verified through unified baseline comparison, ablation experiments and robustness analysis. The conclusion is limited to the current test data set, and external validation in a real organizational environment needs to be carried out in the future.
No takes yet. Share an insight, caveat, or question.
Caitian Zhang (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: