A study of Tibetan text classification based on CBAt hybrid neural network model

Text classification is an important research work in the field of natural language processing, and the emergence of deep learning methods has greatly promoted the development of text classification techniques, which is also effective for Tibetan text classification. This paper proposes a CBAt hybrid neural network based Tibetan text classification model to classify 14,000 texts in Tibetan dataset, using the convolutional results in the traditional CNN model as the input of the BiLSTM model and introducing the Attention mechanism to increase the model's feature extraction of important information of Tibetan texts, so as to improve the classification accuracy. The paper also compares this dataset with traditional machine learning algorithms and a single neural network model, and the experimental results show that the improved hybrid neural network model proposed in this paper performs better in the classification of Tibetan texts.

文本分类是自然语言处理领域的一项重要研究课题，深度学习方法的兴起极大地推动了文本分类技术的进步，这在藏语文本分类中亦显成效。本研究提出了一种基于CBAt混合神经网络的藏语文本分类模型，用以对藏语文本数据集中的14,000篇文本进行分类。该模型以传统卷积神经网络（CNN）的卷积结果作为双向长短时记忆网络（BiLSTM）的输入，并引入注意力机制以提高模型对藏语文本重要信息的特征提取能力，从而提升分类准确率。论文还比较了该数据集与传统的机器学习算法及单一神经网络模型，实验结果表明，本文提出的改进型混合神经网络模型在藏语文本分类方面表现更为优越。