Project Description
Explore external data sources to provide broader insights into wellbeing factors.
Improve wellbeing measurement by finding new ways to look at data.
Identify indicators that capture broader impacts on wellbeing.
Demonstrate potential alternative data sources (e.g. social media) can complement traditional measures.
Data Story
A Specific Domain with a Focused Approach
Goal: Dive deep into a specific domain to uncover insights.
Domain Focus: Community Safety and Domestic Violence
Explore the interaction between safety indicators (e.g., crime data, domestic violence reports) and other wellbeing domains like mental health and housing stability.
Integrate data from domestic violence helplines, police reports, and emergency services to map trends in domestic violence over time. Highlight the role of socioeconomic factors, employment status, and housing situations in predicting vulnerability to domestic violence.
Apply machine learning algorithms to predict high-risk periods or areas for domestic violence, using data from environmental stressors (e.g., economic downturns) and the availability of community support services.
Proposal:
Use natural language processing (NLP) techniques to analyze social media data and news articles to detect emerging wellbeing trends (such as changes in community sentiment or mental health trends related to specific events, such as economic crises or natural disasters). Link these insights to ACT's Wellbeing Framework domains.
Data Collection
Social Media Data: Collect posts from platforms like Twitter, Reddit, or Facebook using APIs (e.g., Twitter API). Define hashtags or keywords related to wellbeing (e.g., #MentalHealth, #Safety, #CommunitySupport).
News Articles: Scrape or use APIs to gather relevant news articles from major Australian or ACT-specific outlets. You can use news APIs like GDELT or Google News API
Correlate social media sentiment with wellbeing indicators (e.g., safety, mental health) by analyzing spikes in negative sentiment during certain periods (e.g., lockdowns) and identifying which domains of wellbeing were most impacted.
Create interactive visualizations that map correlations between domains, such as health, education, and employment, to reveal the interconnectivity of these factors.
Further information can be found in our presentation https://gamma.app/docs/TEAM-GUCK-MAL-EIN-KEKS-Cookiessss-n3q0qzcwgi95tnl
