Kammu, Deborah Ebbu (2022) Research on Negative Post Identification in the Regional Language (Hindi). Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration manual)
Download (3MB) | Preview |
Abstract
Identifying negative posts has become a primary requirement in the current era, where social media plays a significant role in the lives of Internet users. In this context, the paper makes an attempt to determ- ine negative posts in the regional language. Therefore, the aim of the research paper is to identify different methods that can help in determ- ining negative posts in the regional language. In order to fulfil this aim, the research conducts a significant evaluation of other literary works that help in understanding multilingual BERT, ”K-Means and Naive Bayes algorithms”, deep learning approaches and “Recurrent Neural Network- based Approach” as a method of detecting negative posts. Further, par- ticular methodologies are used for creating a design specification tech- nique that helps in elaborating the means of implementing the method in understanding negative posts in Hindi. A post detection technique using BERT training model has been used a best approach in this pro- ject. The main result of this project has been proposed by showing the percentage of accuracy level of different model, SVM and random forest are two most widely used model here.
Actions (login required)
View Item |