News headlines can be a good data source for detecting the news spreading barriers in news media, which may be useful in many real-world applications. In this paper, we utilize semantic knowledge through the inference-based model COMET and sentiments of news headlines for barrier classification. We consider five barriers including cultural, economic, political, linguistic, and geographical, and different types of news headlines including health, sports, science, recreation, games, homes, society, shopping, computers, and business. To that end, we collect and label the news headlines automatically for the barriers using the metadata of news publishers. Then, we utilize the extracted commonsense inferences and sentiments as features to detect the news spreading barriers. We compare our approach to the classical text classification methods, deep learning, and transformer-based methods. The results show that the proposed approach using inferences-based semantic knowledge and sentiment offers better performance than the usual (the average F1-score of the ten categories improves from 0.41, 0.39, 0.59, and 0.59 to 0.47, 0.55, 0.70, and 0.76 for the cultural, economic, political, and geographical respectively) for classifying the news-spreading barriers.

利用 COMET 推理模型和新闻标题的情感，通过采集和标记新闻标题的元数据来检测新闻传播中的文化、经济、政治、语言和地理障碍，并将提取到的语义知识和情感作为特征，相对于传统文本分类方法和深度学习可获得更好的表现（对于文化、经济、政治和地理的十个类别的平均 F1-分数分别从 0.41、0.39、0.59 和 0.59 提高至 0.47、0.55、0.70 和 0.76）。

利用新闻标题研究新闻传播障碍的特征