A Hybrid Dictionary Model for Ethical Analysis
- 43 Downloads
With the adoption of social media and web services, people have become more likely to share opinions on the web about their daily activities. Thus, social networks end up being seen as an opportunity to bend the rules, making things inadmissible in society. This work aimed to design a dictionary template for sentiment analysis applied to unethical behaviors. It also analyses how prepared the machine is for human dialogue, and proposes a hybrid approach combining existing work for dictionary creation and standard conversation recognition in the Internet. Additionally, it also analyzes current policies to remove inappropriate content and proposes some steps to be considered to reduce spread of inappropriate content in the web.
KeywordsUnethical Behaviour Dictionary Customer opinion Sentiment analysis Model
The authors thank: the Ecossistema Negócios Digitais Ltda; ITA—the Brazilian Aeronautics Institute of Technology; and FCMF—The Fundação Casimiro Montenegro Filho, for all their general and financial support, during this investigation.
- 1.Landgrebe, J., Smith, B.: There is no general AI: why Turing machines cannot pass the Turing test, arXiv preprint arXiv:1906.05833 (2019)Google Scholar
- 2.Turing Test. https://en.wikipedia.org/wiki/Turingtest (2019)
- 3.Teste de Turing. https://pt.wikipedia.org/wiki/TestedeTuring (2019)
- 5.Mondal, M., et al.: A measurement study of hate speech in social media. In: Proceedings of the 28th ACM Conference on Hypertext and Social Media, pp. 85–94 (2017)Google Scholar
- 6.IA deve aprender sobre valores sociais para não termos problemas. https://www.tecmundo.com.br/ciencia/145659-ia-deve-aprender-valores-sociais-nao-termos-problemas.htm (2019)
- 7.YouTube explica como funciona o processo de remoção de conteúdo nocivo. https://canaltech.com.br/redes-sociais/youtube-explica-como-funciona-o-processo-de-remocao-de-conteudo-nocivo-148647/ (2019)
- 8.The Four Rs of Responsibility, part 1: removing harmful content. https://youtube.googleblog.com/2019/09/the-four-rs-of-responsibility-remove.html/ (2019)
- 9.YouTube exclui 30 mil vídeo por incitação ao ódio. https://olhardigital.com.br/noticia/youtube-exclui-30-mil-videos-por-incitacao-ao-odio/89927 (2019)
- 10.Sun, Y., et al.: Ernie 2.0: a continual pre-training framework for language understanding, arXiv preprint arXiv:1907.12412 (2019)Google Scholar