Offensive language detection has been well studied in many languages, but it
is lagging behind in low-resource languages, such as Hebrew. In this paper, we
present a new offensive language corpus in Hebrew. A total of 15,881 tweets
were retrieved from Twitter. Each was labeled with one or more of five classes
(abusive, hate, violence, pornographic, or none offensive) by Arabic-Hebrew
bilingual speakers. The annotation process was challenging as each annotator is
expected to be familiar with the Israeli culture, politics, and practices to
understand the context of each tweet. We fine-tuned two Hebrew BERT models,
HeBERT and AlephBERT, using our proposed dataset and another published dataset.
We observed that our data boosts HeBERT performance by 2% when combined with
D_OLaH. Fine-tuning AlephBERT on our data and testing on D_OLaH yields 69%
accuracy, while fine-tuning on D_OLaH and testing on our data yields 57%
accuracy, which may be an indication to the generalizability our data offers.
Our dataset and fine-tuned models are available on GitHub and Huggingface.

在本研究中，我们提供了希伯来语的新冒犯性语言语料库，使用推特上的 15,881 条推文，并由阿拉伯 - 希伯来双语人士对其进行了五个类别（辱骂、仇恨、暴力、色情或无冒犯）的标注，我们使用我们提出的数据集和另一个已发布的数据集对两个希伯来语 BERT 模型（HeBERT 和 AlephBERT）进行了微调，观察到我们的数据与 D_OLaH 相结合可以将 HeBERT 的性能提升 2％。将 AlephBERT 在我们的数据上进行微调并在 D_OLaH 上进行测试，准确率达到了 69％，而在 D_OLaH 上进行微调并在我们的数据上进行测试，准确率达到了 57％，这可能意味着我们提供的数据的通用性。我们的数据集和微调模型已在 GitHub 和 Huggingface 上提供。