Mining social media data of people with rare diseases
summary
Analyzes public Facebook and Twitter posts about rare diseases using Netvizz to pull Facebook data, geolocation-from-profile methods and a Leaflet/GitHub map for Twitter, and Python scripts for sentiment (polarity and subjectivity) and log‑likelihood term ranking. Builds word clouds, engagement buckets and temporal charts that show photos get the most likes, peaks at midday and late night, and Mondays have the most posts. Models health concepts as an ontology in Protégé and publishes it to BioPortal.
AI-written summary · about
chapters (23)
- 0:00Introduction
- 1:22Why rare diseases
- 1:53CERN
- 3:24Google Summer of Code
- 4:34Survey
- 6:06What are rare diseases
- 7:04How rare diseases affect children
- 9:50Objectives
- 10:54Methodology
- 11:45Results
- 13:17Ranking
- 13:39Conclusions
- 14:10Types of posts
- 14:58Temporal analysis
- 16:47Twitter data analysis
- 17:45Map
- 19:44Content analysis
- 20:07Interoperability problem
- 21:59conclusion
- 23:20diversity
- 24:06show your care
- 24:37with stories
- 25:24Thank you
related talks
11mConnecting Researchers to Patient and Person Generated Data
44mAnalyzing social media data with Python
06m11 Years withWearables: Quantitative Analysis of Social Media, Academia, News Agencies, and Lead ...
13mData Nerding in Public Health
20mData-Driven Approach in Treating a Medical Condition
01h 19mOpportunities and risks of big data in disease surveillance.
15mEnhancing Medical Data Processing With Neo4j
16mMedQGraph: TMKGs for AI-Driven Healthcare Insights