There are 3 directories

1. data: contains all the data (emails, train data, test data)
2. scripts: contains scripts to preprocess data, script for making the webApi
3. libshorttext-1.0: library for building the SVM model (a machine learning algorihtm for classification)

Before executing, you might have to re-build the library. (I build it for linux 32 bit) Move to libshorttext-1.0/ directory and execute: make clean make

How to execute

1. create the training an test data Move to scripts directory and execute: "python classifpy"

2. train the model Move to libshorttext-1.0/ directory and execute: "python text-train.py ../data/training_data" 1st argument is the path of training data

3. test the model Move to libshorttext-1.0/ directory and execute: "python text-predict.py ../data/testing_data training_data.model/ results" 1st argument is path of test data, 2nd argument is path to the model, 3rd argument is path to a file for storing results

For more training and testing options, check the README in libshorttext-1.0/

For real time prediction Move to scripts and execute: "python predict.py <PATH_TO_MODEL>" Example: "python predict.py ../libshorttext-1.0/training_data.model/"