M&A Predictive Analysis Project

How to Use

• Clone Repo
• Download the zipped Data from https://drive.google.com/file/d/14HL3TcDUb9mNE1kqSvJGUmQZBTWTbPm8/view?usp=sharing
• Place it into data collection into a folder named NewData

Current repository directory should look like this

• data collection
  • M_A_News.csv
  • NewData
    • [Downloaded Unzipped Data]
• CLEAN_ClassificationNewData.ipynb
• CLEAN_Classification_isM_A.ipynb
• CLEAN_LDAonNews.ipynb
• README.md

From there:

• Run “CLEAN_LDAonNews.ipynb” first to generate “lda.sav” and “cv.sav” in the same folder.
• Make sure the above two models are existing in the “Final Package” folder.
• If you want to find the topics for M&A and general news, run “CLEAN_Classification_isM&A.ipynb”
• For the final classification for 1/3/6/12 month news article, run “CLEAN_ClassificationNewData.ipynb”. This code will generate a sub-folder called “ClassifyNews”. The generated feature table will saved in this folder. The results will print in the stdout console.