Cross-linguistic Information Retrieval Between English and Chinese Based on the Neural Network Model

English and Chinese linguistic conversions are optimal for retrieving information through cross-linguistic content. The information retrieval is maximized using artificial intelligence-based solutions. This article introduces a cross-linguistic information retrieval method to achieve the aforementioned target. The proposed method identifies phrases and information representation formats that differ between Chinese and English. The conversion between the two languages is used to identify their semantics and ensure high information extraction. In this method, the neural network, specifically a Bi-LSTM model, is employed to verify the phrase and representation between different conversion instances. These instances are useful in validating the linguistic structure and semantics of the two languages. The DuIE and InstructIE datasets are used to train and evaluate the model, ensuring its effectiveness in real-world scenarios. This enhances the phrase correlation and word representation to improve and train the neural network. The neural network verifies the retrieval and matching information, as well as phrase semantics, to ensure better cross-linguistic extraction. The article effectively conveys its objective, but it previously lacked specific details about the neural network architecture and datasets used. Including this information enhances the reader’s understanding of the technical contributions. The proposed method is analyzed using a semantic assessment score of 0.81, an information extraction rate of 89.7%, an extraction precision of 86.5%, and a computation efficiency of 7.9 ms per query, demonstrating improved performance in cross-linguistic retrieval tasks.