This repository presents an AI model that can classify SMS messages as either spam or legitimate (ham) with an impressive accuracy of 95%. We employ techniques such as TF-IDF (Term Frequency-Inverse Document Frequency) and Logistic Regression as the classifier to identify spam messages from a given dataset.
The dataset used for this project can be found at the following URL: SMS Spam Collection Dataset.
The dataset contains two main columns:
v1: Contains labels for each SMS message, indicating whether it is "ham" (legitimate) or "spam."v2: Contains the actual SMS message text.
For a detailed implementation and analysis of the SMS spam classification model using Logistic Regression, please refer to the Colab notebook available here: Colab Notebook.
We have developed our SMS spam classification model using Logistic Regression. Logistic Regression is well-suited for binary classification tasks like spam detection, where the goal is to classify messages as either "spam" or "ham."
-
Data Preprocessing: We preprocess the SMS message data, including text cleaning and tokenization.
-
Feature Extraction: We use TF-IDF (Term Frequency-Inverse Document Frequency) to convert the text data into numerical features that can be used by the machine learning model.
-
Model Training: We train a Logistic Regression classifier using the preprocessed and transformed data.
-
Model Evaluation: The model achieved an impressive accuracy of 95%, making it highly reliable in classifying SMS messages correctly.
The Logistic Regression model achieved an accuracy score of 95% on the SMS spam classification task. This high accuracy demonstrates the effectiveness of the model in identifying spam messages.