User-Input-Classifier

---

It classifies the user input in the model into the following three classes:

1. Question - Q
2. Statement - S
3. Chat/Small talk - C

Evaluation

Evaluated the model on Random Forests, Support Vector Machines and K Nearest Neighbors Classifier to determine the best choice of classifier to be used.

The Precision, Recall anf F-1 score metrics are:

1. Random Forest Precision - 0.766835016835 Recall - 0.756756756757 F-1 - 0.757319966062

2. K Nearest Neighbors Precision - 0.726943346509 Recall - 0.702702702703 F-1 - 0.715927750411

3. Support Vector Machines Precision - 0.681081918934 Recall - 0.675675675676 F-1 - 0.667170473205

Requirements

Python3 NumPy Sklearn Pandas

How to Run

Run the MODEL_RUN.py file as follows with the desired user input:

python3 MODEL_RUN.py "Is the earth flat"