Email-Spam-Detection-Using-Logistic-Regression

Project Overview:

This project aims to build a robust email spam detection system using machine learning techniques. The primary objective is to classify emails as spam or not spam with high accuracy. We achieved a 95% accuracy using a logistic regression model, enhanced with feature engineering and TF-IDF vectorization.

Key Features:

Checking and Removing Null Values

• df.isnull().sum()
• isnull function identify the null values in the data frame and sum function sum-up, total number of values in the data frame.

Converting Labels SPAM and HAM into '0' and '1'

• Used LabelEncoder() to convert categorical labels (text) into numeric form.
• LabelEncoder is a class in the sklearn.preprocessing module of scikit-learn.

Data Preprocessing

• Utilized the nltk toolkit to preprocess text data by employing functions such as sent_tokenize and word_tokenize.
• As a result, I enriched my DataFrame with three new columns: number_characters, number_sentences, and number_words.

Model Training

• Model is being trained using logistic regression.

Model Efficiency

• Accuracy is used to calculate the efficiency of the model.

Dependencies

• Python 3.x
• Pandas
• Numpy
• Scikit-learn
• Jupyter
• NLTK
• WordCloud

How to Run

Clone the repository:

https://github.com/raja045/Email-Spam-Detection-Using-Logistic-Regression.git

cd Email-Spam-Detection-Using-Logistic-Regression

Install the required packages

pip install -r requirements.txt

Run the Jupyter notebooks to see the step-by-step process of building the model:

jupyter notebook

• Open SpamDetection.ipynb and run the cells sequentially.