What is it about?
The stigma surrounding the queer community inhibits them from expressing themselves. The queer community often face backlash to reveal about their sexual or gender orientation. To this end, we sought to devise a system that could automate the process of assessing whether a particular tweet divulges information about a person 'coming out of the closet' and quantitatively evaluate the reaction's emotional range.
Featured Image
Why is it important?
It is an inconvenient task to manually collect and annotate data for the purpose of statistical surveys. Our algorithm automates the process of doing so, while also identifying the target subject to whom the sexual/gender identity, and providing a quantitative measure of the degree of sentiment in the subject's reaction. To do this, we used classical natural language processing (NLP) methods, which include topic modeling with LDA to filter out data that isn't pertinent to the paper's objective. Furthermore, after identifying the subject to whom this information is disclosed, we used a classical lexicon-based approach combined with a score function for sentiment analysis. This paper also provides a foundation for work that can be further extended to be generalized across all topics and domains.
Perspectives
The LGBTQIA+ community faces much oppression from the society that hinders their quality of life. We feel that such research would go a long way in collecting data that would further enable analysis into identifying the root cause for this, and potential solutions that could help reduce the discrimination that the community faces. Working on this paper gave us insight into the community's problems, specifically for acceptance. We hope that such work will help rid the world of prejudice against such communities.
Dr. Sanjay Singh
Manipal Institute of Technology, Manipal
Read the Original
This page is a summary of: Out of the Closet: Lexicon Based Sentiment Analysis on Tweets about Homosexuality, October 2019, Institute of Electrical & Electronics Engineers (IEEE),
DOI: 10.1109/tencon.2019.8929379.
You can read the full text:
Contributors
The following have contributed to this page
