Skip to content

A spam classifier is built using distinct tools with the objective of comparing the results of each approach.

License

Notifications You must be signed in to change notification settings

Vitor-Sallenave/Spam-Classifier

Repository files navigation

✉️⚠️ Spam Classifier: different approaches


◼️ First Approach: The first detector was built using the spacy module and the machine learning model Random Forest Classifier.

◼️ Second Approach: The second detector uses the tools offered by Keras to construct a sequential neural network.

◼️ Third Approach: The difference between this method and the other ones is that, besides utilizing Keras and building a neural network, it is also created an customized embedding layer and a tokenizer that transform the messages text into sequences of integers in a fixed length that will be read by the NN.

Releases

No releases published

Packages

No packages published