An IoT-based English translation and teaching using particle swarm optimization and neural network algorithm
Lili Zhang
DOI: https://doi.org/10.1007/s00500-023-09032-9
IF: 3.732
2023-07-31
Soft Computing
Abstract:English language translation teaching plays a crucial role in today’s globalized world, particularly in countries like China, which has seen an increased emphasis on English language proficiency. However, effective teaching and assessment of students’ translation abilities remain significant challenges for educators. To address the challenges of English language translation and teaching, this paper proposes a novel approach that integrates Internet of Things (IoT), Particle Swarm Optimization (PSO), and Neural Network algorithms to assist teachers in evaluating and assessing students’ English translation abilities by enabling them to better fulfill their teaching responsibilities. This paper first describes the computational mechanism of Neural Network algorithm using PSO, which includes the particle coding approach. An application model is designed to assess students’ capacity for learning using English translational instruction. The PSO algorithm gathers information on their learning progress, and fully utilize the analyzed results to develop learning strategies and create teaching materials for various learning types, which is helpful for the quick advancement of English translation teaching. Secondly, this paper develops a translational application model for English language as a practical framework. The proposed method utilizes the upgraded PSO-enabled Neural Network to process English language translation and teaching data by achieving network training in a global optimal state of PSO by reducing training errors. Thirdly, this paper evaluates the effectiveness of the model by comparing the average errors of the training and test samples with different numbers of particles such as 5, 10, and 20. Finally, the results demonstrate high accuracy in measuring students’ translation abilities. Furthermore, the study collects and examines three data sets, specifically English-Foreign Language Translation Corpus (EFLT), English Language Learners’ Corpus (ELLC), and English-Multilingual Parallel Corpus (EMPC), with a detection accuracy of 0.84, 0.97, and 0.65, respectively.
computer science, artificial intelligence, interdisciplinary applications