Cost-Sensitive Learning and Ensemble BERT for Identifying and Categorizing Offensive Language in Social Media

Author(s):  
Fajar Muslim ◽  
Ayu Purwarianti ◽  
Fariska Z Ruskanda
Author(s):  
Vildan Mercan ◽  
Akhtar Jamil ◽  
Alaa Ali Hameed ◽  
Irfan Ahmed Magsi ◽  
Sibghatullah Bazai ◽  
...  

2020 ◽  
Vol 24 (2) ◽  
Author(s):  
Elena Shushkevich ◽  
John Cardiff ◽  
Paolo Rosso ◽  
Liliya Akhtyamova

2019 ◽  
Vol 18 (2) ◽  
pp. 75-83
Author(s):  
Piotr Pawłowski ◽  
Daria Makuch ◽  
Paulina Mazurek ◽  
Adrianna Bartoszek ◽  
Alicja Artych ◽  
...  

AbstractIntroduction. Nowadays, a professional image is an important element of the identity of individual professions. Its formation is a difficult process, dependent on many factors, including the use of new communication channels, such as social media, which in recent years have become a space for expressing social opinion, including those concerning individual professions.Aim. The analysis of the possibilities of using social media in shaping the image of nurses on the Internet.Material and methods. The study was carried out using the comparative method. The subject of the research were websites (fanpages) related to the professional environment of nurses on the social networking site Facebook.com, chosen deliberately according to the adopted criteria.Findings. During the research, differences in the strategy of administering the analyzed websites were identified, depending mainly on the subject matter and purpose of publishing the content. The topicality, visual attractiveness and cohesion were characterized by a high level. The posts appearing on individual websites were written in the language of the recipients, with different publication frequency. The websites created a long-term group of recipients and tried to influence the image of nursing in Poland in a positive way.Conclusions. Content published on social media can affect both the positive and negative image of the nurse in the public opinion. Among the factors that do not affect the image of nurses can be indicated, among others, offensive language of comments and displaying negative traits of nurses. Positive reception guarantees current knowledge in the field of nursing and emphasizing professional competences.


2019 ◽  
Author(s):  
Vijayasaradhi Indurthi ◽  
Bakhtiyar Syed ◽  
Manish Shrivastava ◽  
Manish Gupta ◽  
Vasudeva Varma

2021 ◽  
Vol 1911 (1) ◽  
pp. 012012
Author(s):  
R Geetha ◽  
S Karthika ◽  
Chaluvadi Jwala Sowmika ◽  
Bharathi M Janani

Information ◽  
2021 ◽  
Vol 12 (10) ◽  
pp. 418
Author(s):  
Daniela America da Silva ◽  
Henrique Duarte Borges Louro ◽  
Gildarcio Sousa Goncalves ◽  
Johnny Cardoso Marques ◽  
Luiz Alberto Vieira Dias ◽  
...  

In recent years, we have seen a wide use of Artificial Intelligence (AI) applications in the Internet and everywhere. Natural Language Processing and Machine Learning are important sub-fields of AI that have made Chatbots and Conversational AI applications possible. Those algorithms are built based on historical data in order to create language models, however historical data could be intrinsically discriminatory. This article investigates whether a Conversational AI could identify offensive language and it will show how large language models often produce quite a bit of unethical behavior because of bias in the historical data. Our low-level proof-of-concept will present the challenges to detect offensive language in social media and it will discuss some steps to propitiate strong results in the detection of offensive language and unethical behavior using a Conversational AI.


2021 ◽  
Vol 5 (9) ◽  
pp. 54
Author(s):  
Daryna Dementieva ◽  
Daniil Moskovskiy ◽  
Varvara Logacheva ◽  
David Dale ◽  
Olga Kozlova ◽  
...  

We introduce the first study of the automatic detoxification of Russian texts to combat offensive language. This kind of textual style transfer can be used for processing toxic content on social media or for eliminating toxicity in automatically generated texts. While much work has been done for the English language in this field, there are no works on detoxification for the Russian language. We suggest two types of models—an approach based on BERT architecture that performs local corrections and a supervised approach based on a pretrained GPT-2 language model. We compare these methods with several baselines. In addition, we provide the training datasets and describe the evaluation setup and metrics for automatic and manual evaluation. The results show that the tested approaches can be successfully used for detoxification, although there is room for improvement.


Author(s):  
Alessandro Maisto ◽  
Serena Pelosi ◽  
Simonetta Vietri ◽  
Pierluigi Vitale

2019 ◽  
Author(s):  
Gabriel Florentin Patras ◽  
Diana Florina Lungu ◽  
Daniela Gifu ◽  
Diana Trandabat

Sign in / Sign up

Export Citation Format

Share Document