arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.04616cs.CL
收藏

希望在资源不足的卡纳达语中进行语音检测

Hope Speech detection in under-resourced Kannada language

Indian Institute of Information Technology Tiruchirappalli（印度信息技术学院蒂鲁吉拉帕利分校）；ULTRA Arts and Science College, Madurai Kamaraj University（马杜赖卡马拉吉大学 ULTRA 文理学院）；Kongu Engineering College（孔古工程学院）；Sultan Idris Education University（苏丹依德里斯教育大学）；Mu Sigma Inc.（穆西格玛公司）；Insight SFI Research Centre for Data Analytics, Data Science Institute, National University of Ireland Galway（爱尔兰国立高威大学数据科学研究所 Insight SFI 数据分析研究中心）

Adeep Hande, Ruba Priyadharshini, Anbukkarasi Sampath, Kingston Pal Thamburaj, Prabakaran Chandran, Bharathi Raja Chakravarthi

阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

摘要：

近年来，通过消除社交媒体平台上的粗俗，攻击性和激烈的评论，已经开发了许多方法来监控负面情绪的传播。然而，有相对较少的研究集中在拥抱积极性，加强在线论坛中的支持性和令人放心的内容。因此，我们建议创建一个英语卡纳达语希望语音数据集，KanHope和比较几个实验来基准数据集。该数据集由6，176个用户生成的评论组成，这些评论混合了从YouTube上抓取的卡纳达语代码，并手动注释为带有希望的语音或没有希望的语音。此外，我们还介绍了DC-BERT4HOPE，这是一种双通道模型，它使用KanHope的英文翻译进行额外的训练，以促进希望语音检测。该方法的加权F1得分为0.756，优于其他模型。虽然如此，KanHope旨在鼓励卡纳达语的研究，同时广泛促进研究人员对鼓励，积极和支持的在线内容采取务实的方法。

英文摘要：

Numerous methods have been developed to monitor the spread of negativity in modern years by eliminating vulgar, offensive, and fierce comments from social media platforms. However, there are relatively lesser amounts of study that converges on embracing positivity, reinforcing supportive and reassuring content in online forums. Consequently, we propose creating an English-Kannada Hope speech dataset, KanHope and comparing several experiments to benchmark the dataset. The dataset consists of 6,176 user-generated comments in code mixed Kannada scraped from YouTube and manually annotated as bearing hope speech or Not-hope speech. In addition, we introduce DC-BERT4HOPE, a dual-channel model that uses the English translation of KanHope for additional training to promote hope speech detection. The approach achieves a weighted F1-score of 0.756, bettering other models. Henceforth, KanHope aims to instigate research in Kannada while broadly promoting researchers to take a pragmatic approach towards online content that encourages, positive, and supportive.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑