new_data

This study used a benchmark dataset, applying different embedding like LASER and FastText to capture contextual information, which was combined to create a new hybrid embedding. This hybrid embedding was fed to machine-learning (ML) and deep learning (DL) classifiers. ML classifiers consist of three ensembles: bagging utilizes Random Forest (RF), boosting with Light Gradient Boosting Machine (LGBM) and Xtreme Gradient Boosting (XGB), and stacking with Support Vector Machine (SVC) and XGB as base learners and Logistic Regression (LR) as the Meta classifier. DL classifiers such as fully connected network (FCN), Convolutional neural network (CNN), long short-term memory (LSTM), and gated recurrent unit (GRU) have been employed. The model's performance was assessed through independent set and k-fold testing with 5 and 10 folds, using evaluation metrics such as Accuracy, Recall, Precision, and F1 score. In experiments, DL classifiers have outperformed ML classifiers regarding accuracy score. The proposed model outperformed previous benchmark studies, achieving an Accuracy of 0.86, Recall of 0.901, Precision of 0.866, and F1 score of 0.883 in the independent set test. This method contributes to NLP research by showcasing the utility of a hybrid embedding approach.

本研究采用基准数据集，应用LASER与FastText等不同嵌入方法捕获上下文信息，并将各嵌入方法的结果融合以构建新型混合嵌入（hybrid embedding）。将该混合嵌入输入至机器学习（Machine Learning, ML）与深度学习（Deep Learning, DL）分类器中。机器学习分类器涵盖三类集成学习方案：套袋（bagging）集成采用随机森林（Random Forest, RF），提升（boosting）集成采用轻量级梯度提升机（Light Gradient Boosting Machine, LGBM）与极限梯度提升树（Xtreme Gradient Boosting, XGB），堆叠（stacking）集成则以支持向量机（Support Vector Machine, SVC）与XGB作为基学习器，以逻辑回归（Logistic Regression, LR）作为元分类器。深度学习分类器则采用了全连接网络（Fully Connected Network, FCN）、卷积神经网络（Convolutional Neural Network, CNN）、长短期记忆网络（Long Short-Term Memory, LSTM）以及门控循环单元（Gated Recurrent Unit, GRU）。本研究通过独立测试集与5折、10折交叉验证对模型性能进行评估，所采用的评价指标包括准确率（Accuracy）、召回率（Recall）、精确率（Precision）与F1值（F1 score）。实验结果表明，深度学习分类器在准确率指标上优于机器学习分类器。所提出的模型性能优于既往基准研究，在独立测试集测试中实现了0.86的准确率、0.901的召回率、0.866的精确率以及0.883的F1值。本方法通过验证混合嵌入方案的有效性，为自然语言处理（Natural Language Processing, NLP）研究提供了有益参考。