Data for "Based on Data Balancing and Model Improvement for Multi-Label Emotion Recognition"

Dataset Description Data for “Based on Data Balancing and Model Improvement for Multi-Label Emotion Recognition” This repository contains the data supporting our study on multi-label emotion recognition. The dataset and materials are shared under the CC-BY 4.0 license. Core Dataset balanced_emotion_dataset.csv – The final balanced multi-label sentiment dataset used for training and evaluation of our model. This file was renamed from the original final_balanced_df_output.csv. Columns: text, sentiment (a list of emotion labels for the text). Data for Figures fig2_balanced_label_counts.csv – Data for Figure 2. Contains the counts of each of the 28 emotion labels in our final balanced dataset. Columns: Sentiment Labels, Counts. training_history.csv – Data for Figures 6 & 7. Contains the training history log with loss and accuracy for each epoch on both the training and validation sets. Columns: epoch, accuracy, loss, val_accuracy, val_loss. Source Code model.ipynb – The original Jupyter Notebook containing the full code for data processing, model training, and evaluation, from which the above data files were derived.数据集说明 本数据集对应《基于数据均衡与模型优化的多标签情感识别（Multi-Label Emotion Recognition）》研究所用数据 本仓库存储了支撑我们开展多标签情感识别研究的相关数据。本数据集及配套材料采用CC-BY 4.0许可协议进行共享。 核心数据集 balanced_emotion_dataset.csv：用于模型训练与评估的最终均衡化多标签情感数据集，该文件由原文件名final_balanced_df_output.csv重命名而来。 字段说明：text（文本内容）、sentiment（该文本对应的情感标签列表）。 绘图配套数据 fig2_balanced_label_counts.csv：图2的配套数据，包含最终均衡化数据集中28种情感标签各自的计数统计结果。 字段说明：Sentiment Labels（情感标签）、Counts（标签计数）。 training_history.csv：图6与图7的配套数据，包含训练集与验证集每一轮训练的损失值与准确率的训练日志记录。 字段说明：epoch（训练轮次）、accuracy（训练准确率）、loss（训练损失）、val_accuracy（验证集准确率）、val_loss（验证集损失）。 源代码 model.ipynb：包含完整的数据处理、模型训练与评估代码的原始Jupyter Notebook文件，上述所有数据集文件均由此脚本生成。