cmsptcp/tsmp

Twitter based stock market prediction using Pyspark, project for Big Data PW 2020L

tsmp

Twitter based stock market prediction, project for Big Data PW 2020L

Used technologies:

• Pyspark, Spark.SQL, Spark.ML, SparkNLP
• Pandas & Numpy
• Docker

Files:

• prototype/tsmp.ipynb contains version with calculations for 15 minutes stock ticker_interval
• prorotype/data_download contains basic python scripts used to download data from Twitter and Yahoo Finance - please remember to use your own key and bearer token for Twitter download (to be set in app_config)
• prototype/sample_data contains example data downloaded from Twitter and Yahoo Finance