Using Natural Language Processing and Machine Learning techniques to automatically detect cyberbullying.

使用自然语言处理和机器学习技术自动检测网络欺凌。

• 批准号: 2253445
• 金额: --
• 依托单位: University College London
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2019
• 资助国家: 英国
• 起止时间: 2019 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2253445
• 关键词: Using; Natural; Language; Processing; Machine

Bullying affects millions of children throughout the world each year. The Department for Education states that in the UK alone, 1 in 6 children aged 10-15 have reported being bullied, and 7% of these children have experienced some form of cyberbullying and cyberthreats made to them. Automatic cyberbullying detection is a task of growing interest, particularly in the Natural Language Processing and Machine Learning communities. As a result, it can be used to prevent individuals from receiving harmful online content in social networks, therefore aiding in the reduction of the incidence of cyberbullyi ng. This is a rapidly growing field of work, and I am excited to be contributing to it, especially since the state -of-the-art in automatic detection is developing in ways that we never imagined possible. However, one area continues to be overlooked and that is machine translation, especially in the Arabic language. This is due to the lexicology and sentiment analysis in the various dialects; MSA (modern standard Arabic) that is used for reading and writing official texts is not the same that is used forspeaking and writing in social media networks. Using my background in both languages, I aim to deepen the understanding of how machine translation operates in Arabic, and how hate speech detection and hence cyberbullying detection and deterrence can be implemented on a wider scale. Youth from many backgrounds can benefit from the results that this research will bring. 1 am aware of the high level of knowledge and experience I will need to gain within many areas: some of which are machine learning, Python programming, deep learning, sentiment analysis, lexicology, and better research methods, and I am looking forward to rising to the challenge. I am hoping to make use of all the state-of-the-art technologies that we have available at UCL as well as collaborating with the leading innovators in this field to make an impact in this field of research.

欺凌行为每年影响全世界数百万儿童。英国教育部表示，仅在英国，就有六分之一的 10 至 15 岁儿童报告遭受过欺凌，其中 7% 的儿童经历过某种形式的网络欺凌和网络威胁。自动网络欺凌检测是一项越来越受关注的任务，特别是在自然语言处理和机器学习社区中。因此，它可以用来防止个人在社交网络中接收有害的在线内容，从而有助于减少网络欺凌的发生率。这是一个快速发展的工作领域，我很高兴能为它做出贡献，特别是因为自动检测的最先进技术正在以我们从未想象过的方式发展。然而，有一个领域仍然被忽视，那就是机器翻译，尤其是阿拉伯语翻译。这是由于各种方言的词汇学和情感分析；用于阅读和书写官方文本的 MSA（现代标准阿拉伯语）与社交媒体网络中用于口语和写作的语言不同。利用我的两种语言背景，我的目标是加深对阿拉伯语机器翻译如何运作的理解，以及如何在更广泛的范围内实施仇恨言论检测以及网络欺凌检测和威慑。来自不同背景的年轻人可以从这项研究带来的成果中受益。 1 我意识到我需要在许多领域获得高水平的知识和经验：其中一些是机器学习、Python 编程、深度学习、情感分析、词汇学和更好的研究方法，我期待着迎接挑战。我希望利用伦敦大学学院拥有的所有最先进的技术，并与该领域的领先创新者合作，在该研究领域产生影响。