Introduction 

added_eScience_2019_SMM.pptx

Geting start

• Register on https://socialmediamacroscope.org and start trying out the tools
• Code base: 
  • https://opensource.ncsa.illinois.edu/bitbucket/projects/SMM (for development)
  • https://github.com/IllinoisSocialMediaMacroscope (mirrored from bitbucket) 

Contributing to existing tools

• SMILE

• SMILE is a comprehensive social media data collecting and analytics platform.

• Defining new social media platform data source

  • how to authenticate and authorize
  • what's the data structure like
  • what is the platforms policy on data collection and ownership
  • learn GraphQL and define your own schema

• Adding new analytics metrics

  • currently we have Sentiment analysis, NLP Preprocessing, Topic modeling, Automated phrase detection, Name Entity Recognition, and Twitter Network analysis
  • you can bring in your own analyses, ideally in Python, but can be any programming languages.
  • Alternatively,  in each of the above mentioned analyses, if you have specific algorithms or innovated ways to perform that analysis, you are welcomed to bring in those
  • Useful links:
    • Guidance for writing configuration file in SMILE
    • Template for submitting Python algorithm
    • SMILE endpoints

• BAE

• PAHO COVID19 dashboards

• This is the newest project we are working on currently!

Other aspects that you could contribute to:

• UI/UX design

• Testing and Monitoring

• Exploring new computation platforms and resources

  • Other cloud computing platform capability other than AWS; for example Azure since we have allocation for that
  • Container orchestration - one of the direction we are exploring is to dockerize all the components of the tools,  and we plan to scale it up using kubernetes
  • Exploring the capability of existing analytics and integrate to the tools: such as Google Trend API, IBM watson, Tableau and etc

• Community Engagement:

  • How to advertise SMM project and tools to the community; How to find our target audience and satisfy their needs
  • Workflow or standards to include new data source and algorithms

Writing your own tools

• Could be anything novel that utilizing social media data and analyses; preferably using opensource libraries and algorithms that associates with academic publication

• Need to have a Graphic Interface

• Plan ahead: Need to communicate with Hubzero (where all the tools have been hosted) to decide if the technology stack you plan to use is compatible with their platform