Description
For this challenge you will create an AI algorithm to detect spam emails from a csv file. Similar to the spam email detection part 1, except this time you will not hardcode the code the text runs through to detect whether its spam or not. Instead you will use keras or tenserflow to create a model that reads the csv file by searching for the "Body Paragraph" column and reading it to then predict whether or not that text was an indication of spam or real. To help start you off, we will say the hardest part of this challenge is processing the data to feed to a model, but once you do that, everything falls in place. If you are new to the field of machine learning and would still like to participate we encourage you to watch some videos or articles from our tutorial page showing how to process data and create a working model. The dataset we provided is from the Kaggle database. You are allowed to add any data of your own.
Input
Your binary classification model will read a .csv file that contains the input and output (body paragraph and real value) Unlike part I there is no subject or email address given in this challenge. The real value is "spam" (1) or "real" (0). The data is added in the examples section below. Test cases will also given in csv format.
Output
Your binary classification model should return a boolean value, True (1) or False (0). True (1) means that the function is saying the email is spam. False (0) means that the function is real, or not spam.