Power Semantic Text Relation Extraction Method Based on Data Center

作者：Jiali Sun, Cong Hu, Cuicui Zhang, Peng Wang, Qi Sun, Ruixuan Lu · 发表于：2022 IEEE 10th Joint International Information Technology and Artificial Intelligence Conference (ITAIC) · 年份：2022 · DOI：10.1109/itaic54216.2022.9836896 · 被引用次数：1 · 研究领域：Topic Modeling、Advanced Text Analysis Techniques、Natural Language Processing Techniques

At present, traditional deep learning-based relation extraction methods are difficult to extract in complex contexts, and do not consider the impact of non-target relations in context on relation extraction. In response to this problem, this paper proposes a control input long short-term memory network. CI-LSTM, this network adds an input control unit composed of an attention mechanism and a control gate valve unit to the traditional LSTM. The control gate valve unit can perform key learning on key positions according to the control vector. The different features of the input are calculated. In this paper, the syntactic dependency relationship is finally selected to generate the control vector and the relationship extraction model is constructed through experiments. Better performance in complex contexts.