https://github.com/sukanyadutta52/sentiment-analysis

An Analysis of How Machine Perceives Women and How Women Feel about Themselves As a Result of This Perception: Sentiment Analysis
https://github.com/sukanyadutta52/sentiment-analysis

flair matplotlib nltk-library pandas regular-expression sentiment-analysis spacy textblob vader-sentiment-analysis women-beauty-standard

Last synced: 16 days ago
JSON representation

An Analysis of How Machine Perceives Women and How Women Feel about Themselves As a Result of This Perception: Sentiment Analysis

• Host: GitHub
• URL: https://github.com/sukanyadutta52/sentiment-analysis
• Owner: sukanyadutta52
• Created: 2024-05-05T19:41:51.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-10-15T17:38:17.000Z (2 months ago)
• Last Synced: 2024-10-25T06:01:21.882Z (about 2 months ago)
• Topics: flair, matplotlib, nltk-library, pandas, regular-expression, sentiment-analysis, spacy, textblob, vader-sentiment-analysis, women-beauty-standard
• Language: Jupyter Notebook
• Homepage:
• Size: 187 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Sentiment Analysis

## An Analysis of How Machine Perceives Women and How They Feel about Themselves As a Result of This Perception

* Objective : The project highlights how unrealistic beauty standards, perpetuated by media and toys like Barbie, negatively impact women's self-esteem, body image, and mental health, leading to issues like anxiety, depression, and eating disorders. The research aims to explore how cognitive processes influence these harmful effects and how artificial intelligence, through sentiment analysis, could help address emotions and sentiments related to these challenges.

* Procedure : VADER, a lexicon-based sentiment analysis tool for social media, shows that punctuation, capitalization, and modifiers impact sentiment analysis accuracy, achieving better results than human raters. In the corpus analyzed, VADER indicated 71% neutral, 20% positive, and 9% negative sentiment, with surprising positivity. Flair, using a character-level LSTM, classified the same text as 98% negative, while frequency analysis revealed that words like "beauty," "women," "skin," and "standards" are most frequently repeated, highlighting societal focus on women's appearance and media-driven beauty standards.

* Conclusion : While words like "perfect," "clean," "ideal," and "confidence" may appear positive, in the context of the corpus, they often reflect negative emotions, as women undergo countless procedures daily to achieve these ideals. As data grows more complex, multiple topics emerge, making it possible to create various scenarios from vast amounts of unstructured data, especially in the era of big data, where there is a constant desire to extract as much information as possible.