Create, Train and Export a ML Model

[Minitour]

In order for us to have a good understanding of the outcome of the tweet we need to see if the tweet is positive or negative. For this we will need to use some machine learning techniques.

For this task we'll need to create a new ML model which will be trained off of real tweets.

See this blog post: https://www.twilio.com/blog/2017/12/sentiment-analysis-scikit-learn.html

As well as this repository: https://github.com/lesley2958/natural-language-processing

Finally for saving and loading the model: https://machinelearningmastery.com/save-load-machine-learning-models-python-scikit-learn/

The task

• Create a python script called nlp_train.py which will read two files (positive and negative tweets) and created a trained model. This script must also write the model to disk. The name of the model will be sentimental_score_nlp.dat

• Create a python script called nlp_test.py. This script will load the model created in the training and load it into memory. The script will also contain a function which is called test(str) will return the sentimental score of a given string which varies between 0.0 and 1.0.

[Minitour]

@TomerGoldfeder don't forget to provide an example on how to "import" the script that you wrote and how to use the functions.