This project utilizes a Python Script to detect if news (generated from .csv file) is Real or Fake.
With the use of sklearn, the goal is to build a TfidfVectorizer on the dataset. Then, compute a PassiveAggressive Classifier and fit the model. The accuracy score and confusion matrix will tell how well the model detects fake news.
USEFUL TERMS:
TF (Term Frequency): The number of times a word appears in a document.
IDF (Inverse Document Frequency): Words that can occur many times a document, but are of little importance (such as "is", "of", or "that").
Passive Aggressive algorithms are online learning algorithms. This algorithm remains passive for a correct classification outcome, and turns aggressive in the event of a miscalculation, updating and adjusting. Its purpose is to make updates that correct the loss, causing very little change in the norm of the weight vector.
OUTPUT
