This project analyzes the portrayal of Artificial Intelligence (AI) in Philippine mainstream media, providing insights into public perception, sentiment, and key discussions surrounding AI's development and adoption in the country. This was a group project, and our instructor, a Data Scientist at Rappler News Media, provided valuable guidance and insights, connecting our academic work to real-world media analysis.
Understanding public perception of AI is crucial for effective policy-making, technology adoption, and addressing societal concerns. This project aims to systematically analyze how AI is framed in local media, identifying prevailing sentiments and thematic trends.
The project involved developing a comprehensive Natural Language Processing (NLP) pipeline to process and analyze large volumes of text data from Philippine media sources. This pipeline enabled the extraction of sentiment, identification of key themes, and analysis of framing around AI.
These features highlight the core data science and NLP skills applied in this project:
- Sentiment & Trend Analysis: Identified overall sentiment (positive, negative, neutral) towards AI and tracked its evolution over time (2020-2025).
- Thematic & Framing Analysis: Analyzed common topics, language, and whether AI is predominantly presented as a risk or an opportunity.
- Adoption & Policy Exploration: Explored media discussions related to AI readiness, regulatory frameworks, and key influencers in the Philippine AI landscape.
- Robust Data Collection: Utilized web scraping tools to efficiently gather thousands of media entries from diverse local sources.
- NLP & Machine Learning: spaCy, Machine Learning, Lemmatization, Tokenization, Text Vectorization, Topic Modeling, OpenAI API
- Web Scraping: Selenium, Requests, Beautiful Soup
- Programming Language: Python
This project provides valuable insights into the media's role in shaping public discourse on AI in the Philippines, highlighting areas of concern, opportunities, and the overall perception of this transformative technology. The findings can inform stakeholders on communication strategies and policy development.
Building upon the current insights, potential future enhancements for this project include:
- Expanding Data Sources: Incorporating social media and academic publications to broaden the scope of analysis.
- Granular Sentiment Analysis: Developing more nuanced sentiment analysis techniques for deeper insights.
- Interactive Dashboards: Creating interactive dashboards for real-time trend monitoring and exploration.
- Live Demonstration: Exploring the deployment of an interactive component or visualization (e.g., a web application showcasing sentiment trends or topic models) to provide a dynamic demonstration of the project's capabilities.
- Quantifiable Impact: Further refining the "Impact & Results" section by adding specific, measurable metrics (e.g., "achieved X% accuracy in sentiment classification," "identified Y distinct themes with Z confidence") to highlight the project's effectiveness more precisely.
[FILL THIS SECTION: GitHub Link]
- Email: [email protected]
- LinkedIn: https://www.linkedin.com/in/mljosh/
- GitHub: https://github.com/mjoshua97241