Skip to content

PeerProducedResearch/quantified-self-forums

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

83 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

quantified-self-forums

Self-Quantified is a Community website of self-trackers, self-researchers interested in personal science. It encourages all types of researches to post about their research, ask for advices and explore different methods of self-tracking through wearables.

  • Posts by categories with various tags; ‘data’, ‘tools’, ‘diet’, ‘conference’, ‘food’…

QS wordcloud

Online Dashboard

If you want to have closer look to the results click INTERACTIVE DASHBOARD HERE

To run locally

Please install the project dependencies run pip install -r requirements.txt

$ pip install -r requirements.txt
  • 2nd Please install Pycharm , you can choose PyCharm Community Edition, it's free.

  • 3rd CE example: run on Pycharm terminal:

$ streamlit run webapp/app.py

Expected Results

  • The aim of this project is working to understand the community forum by conducting Data Analysis and Natural Language Processing (NLP) of the community’ interactions.
  • Providing transparent analysis of the human behaviour in communication and their patterns of networking to improve occurring and future projects in community and personal science.

Preliminary results

  • Clustering PCA Model showing type of user engagement
  • LDA MODEL, Topic Modelling showing different posts topics
  • Named Entity Recognition, products, organisation, person
  • Text Classification
  • Network Social Analysis graph

Technical Environment

  • Platform: Python, CorTexT Platform, Pycharm, GEPHI-0.9.2 software for network analysis.
  • PythonLibraries used for pre-processing data: NLTK, SPACY, GENISM.
  • Libraries for modelling, data viz plLDAvis, sklearn, matplotlib, wordcloud, plotly, seaborn, requests, beautiful soup

About

analyses of the QS forum posts

Resources

License

Stars

0 stars

Watchers

1 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors