To run this example:

1. Download the files
2. Must have NODE JS installed
3. Run npm install

Machine Learning SMS

A Machine Learning application which understands if a message is Spam or not.

1) Build your data model

Use the text file sms_dataset.txt with the data from the customer.

Congratulations, you've won a prize!	Spam
Your bank account needs verification	Spam
Win big today! Click here!	Spam
See you at the office tomorrow!	Not Spam
Dinner at 7 pm?	Not Spam
Hey, are we still on for lunch?	Not Spam

2) Run the trainer

Run node model_trainer.js to read the text content and generate a new file called spamClassifier.json This JSON file containsd the training for messages to analyse.

3) Test some messages based on the learning

Run node search_for_spam.js with some messages in the array provided inside the code

const testMessages = [
    "Congratulations, you've won a prize!",         // Spam (expected)
    "Your account has been compromised, act now!",  // Spam (expected)
    "Let's meet for coffee tomorrow at 10",         // Not Spam (expected)
    "Reminder: Your meeting is scheduled at 3 pm",  // Not Spam (expected)
    "Click here to claim your free gift!",          // Spam (expected)
    "Are we still on for dinner tonight?",          // Not Spam (expected)
];

The output will tell you which one of these are spam and which are not.