Design and Implementation of a Modern Chinese History QA System Based on Knowledge Graphs

To address the challenge of quickly obtaining accurate answers while exploring modern historical knowledge, this study develops an intelligent question-answering system based on knowledge graphs, focusing on modern Chinese history. The system leverages the Robustly Optimized BERT Approach (RoBERTa) model for word embedding construction and integrates an adversarial training mechanism to enhance robustness. BiLSTM is employed to capture the contextual features of input text, while Conditional Random Fields (CRF) are used to generate the optimal prediction sequence. Experimental results demonstrate that the model achieves an accuracy of 92.2%, a recall of 94.2%, and an F1-score of 93.0% on the modern history dataset, significantly outperforming other approaches.