Instagram, a social media platform, has in the vicinity of 2 billion active
users in 2023. The platform allows users to post photos and videos with one
another. However, cyberbullying remains a significant problem for about 50% of
young Indonesians. To address this issue, sentiment analysis for comment
filtering uses a Support Vector Machine (SVM) and Easy Data Augmentation (EDA).
EDA will augment the dataset, enabling robust prediction and analysis of
cyberbullying by introducing more variation. Based on the tests, SVM
combination with EDA results in a 2.52% increase in the k-Fold Cross Validation
score. Our proposed approach shows an improved accuracy of 92.5%, 2.5% higher
than that of the existing state-of-the-art method. To maintain the
reproducibility and replicability of this research, the source code can be
accessed at uns.id/eda_svm.

通过使用支持向量机（SVM）和简单数据增强（EDA）进行情感分析的评论过滤，以解决 Instagram 中青少年印尼人群 50% 遭受的网络欺凌问题，测试结果显示 SVM 与 EDA 相结合可使 k-Fold 交叉验证得分提高 2.52%，我们的方法准确率为 92.5%，比现有的最先进方法提高了 2.5%。

在网络欺凌情感分析中的简易数据增强

Easy Data Augmentation in Sentiment Analysis of Cyberbullying

Text augmentation is a technique for constructing synthetic data from an
under-resourced corpus to improve predictive performance. Synthetic data
generation is common in numerous domains. However, recently text augmentation
has emerged in natural language processing (NLP) to improve downstream tasks.
One of the current state-of-the-art text augmentation techniques is easy data
augmentation (EDA), which augments the training data by injecting and replacing
synonyms and randomly permuting sentences. One major obstacle with EDA is the
need for versatile and complete synonym dictionaries, which cannot be easily
found in low-resource languages. To improve the utility of EDA, we propose two
extensions, easy distributional data augmentation (EDDA) and type specific
similar word replacement (TSSR), which uses semantic word context information
and part-of-speech tags for word replacement and augmentation. In an extensive
empirical evaluation, we show the utility of the proposed methods, measured by
F1 score, on two representative datasets in Swedish as an example of a
low-resource language. With the proposed methods, we show that augmented data
improve classification performances in low-resource settings.

通过提出易分布数据增强（EDDA）和类型特定的相似词替换（TSSR）两种扩展方法，利用语义词上下文信息和词性标签进行词替换和增强，我们展示了这些方法在低资源语言的两个代表数据集上通过 F1 分数度量的实用性，证明了增强数据可以在低资源环境中提高分类性能。