2020 Analysis and Application of Big Data final project
Dataset used: Beijing PM2.5 Data Set

Goal: To figure out the correlation of the weather conditions and the pm2.5 indices.
Several classifying methods(Random Forest/Baye's/Logiatic Regression) have been applied.

• Data Preprocessing
  • Missing values replaced by arithmetic mean of the rest of the values.
  • Remove useless/redundant data.
  • Encoding categorical data.
  • Data binning('excellent', 'good', 'light', 'moderate','heavy',and 'severe')
  • Data normalization(min-max normalization).
  • Splitting the dataset into training set and test set.