This project tackles the issue of spam emails by automatically classifying incoming emails as either spam or legitimate (often referred to as "ham").
Who it's for:
Individuals: Anyone who receives a significant amount of email and wants to reduce the clutter and frustration caused by spam. Organizations: Businesses, schools, and other institutions that deal with a high volume of email can benefit from improved spam filtering to: Protect users from phishing attempts and malicious content. Increase productivity by reducing time spent sorting through spam. Reduce server load by filtering out unwanted emails. By automatically classifying emails, this project can help users save time and effort managing their inboxes, and organizations can improve security and efficiency.
- Download he Dataset for Custom Training
- https://in.docworkspace.com/d/sIL6JkZqEAfP307IG?sa=cl
- Python 3.5+
- Jupyter notebook/google collab
-
install numpy
pip install numpy as np
-
install pandas
pip install pandas
-
install sklearn
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression
from sklearn.metrics
To run tests, run the following command
input_mail = [""]
input_data_features = feature_extraction.transform(input_mail)
prediction = model.predict(input_data_features)
-
on Training data
0.9670181736594121
-
on testing data
0.9659192825112107
Mutthuluri varun kumar
zeel Shah
G SUBHAM KUMAR
K MOHAN SAI VARDHAN