/Information-Retrieval

Primary LanguageJupyter Notebook

Information-Retrieval

Front end - Heroku Heroku Demo

Back end - Solr on Free Trial VM (until 20 May)

Files and Folder Structure

  • Combined.csv (Combined data from Dumps Folder in csv)
  • Combined.json (Combined data from Dumps Folder in json)
  • Readme.md
  • Classification
    • Files used and outputs for ML Classification
  • Crawler
    • Source code of Crawler
  • Dumps
    • Raw dumps from Crawler
  • Hand Labelled
    • Labelled dumps
  • Site
    • Front end source code (PHP)

Runtime Instruction

Back End

  1. Setup a Solr Server
  2. Start a Cloud using ./bin/solr start -c
  3. Create a collection using ./bin/solr create -c CovidWatch -s 2 -rf 2
  • This creates a Solr Collection with 2 Shards and 2 Replica
  1. Index the collection using ./bin/post -c CovidWatch /path/to/data/folder
  • For an example, I have cloned this repository at my home folder
  • ./bin/post -c CovidWatch ~/Information-Retrieval/Classification/With general news dataset/Combined Predicted.csv
  1. Solr server is ready to be queried

Front End

  1. Point the Document Root to Site Folder
  2. Change the IP at index.php line 104 and query.php line 71 to your Solr server's IP
  3. Front end is ready