Queensight-AI

Queensight-AI: Gender Equality Network Analysis

Comment

Queensight-AI Project

Inspiration

We were inspired by the persistent gender inequality issues in various sectors and noticed how social media conversations shape public perception. We wanted to create a tool that could analyze these conversations to identify influential voices, emerging topics, and community structures to better understand and address gender equality challenges.

What it does

Queensight-AI is a specialized AI tool that analyzes social networks focused on gender equality topics. It maps connections between users discussing gender issues on platforms like Twitter and Reddit, identifies influential voices, detects key topics, and visualizes conversation networks. Users can interact through a conversational interface to gain insights about the gender pay gap and other equality issues.

How we built it

  • Python Backend: Using NetworkX for graph analysis, Streamlit for the interface
  • AI Integration: Connected to Azure OpenAI's o3-mini model for intelligent conversation
  • Data Collection: Built tools to gather data from Twitter and Reddit APIs
  • Graph Visualization: Created interactive network visualizations to map conversations
  • Conversational UI: Designed a simple chat interface for easy interaction

Challenges we ran into

  • Integrating the Azure OpenAI API required careful error handling
  • Building a responsive chat interface while maintaining graph analysis capabilities
  • Balancing specialized gender equality analysis with general conversational abilities
  • Managing API keys and credentials securely through environment variables
  • Ensuring the tool could handle both technical analysis and user-friendly explanations

Accomplishments that we're proud of

  • Created a specialized AI tool focused on an important social issue
  • Successfully integrated advanced graph analysis with conversational AI
  • Built a clean, user-friendly interface for complex social network analysis
  • Developed a system that can visualize influence networks and community structures
  • Made complex social network analysis accessible through natural language

What we learned

  • How to effectively analyze social media conversations as network structures
  • Techniques for identifying influential voices in online conversations
  • Integration patterns for combining specialized analysis tools with conversational AI
  • Methods for secure credential management in web applications
  • Design patterns for creating accessible technical tools

What's next for Queensight-AI

  • Expanding data collection to more platforms and languages
  • Adding sentiment analysis to measure emotional responses to gender equality topics
  • Creating more advanced visualization options for deeper network insights
  • Building recommendation features to connect like-minded advocates
  • Developing partnership features to help organizations coordinate equality initiatives

Built With

Share this project:

Updates