polifonia-project/musow-pipeline: First version of the pipeline for music resources discovery

The pipeline includes the following components

• a preliminary analysis on a corpus of music resources descriptions, to extract relevant bigrams
• a Twitter API searching for tweets relevant to bigrams
• a classifier of relevant tweets, based on LR and TF-IDf
• a web scraper of web resources referenced in tweets
• a classifier of relevant music web source, based on LR and TF-IDf
• an output table including recommended resources and metadata