https://github.com/kane-dylan/spamshield
SpamShield
https://github.com/kane-dylan/spamshield
mechine-learing numpy pandas python scikit-learn scipy streamlit
Last synced: about 1 month ago
JSON representation
SpamShield
- Host: GitHub
- URL: https://github.com/kane-dylan/spamshield
- Owner: Kane-dylan
- Created: 2025-01-25T08:03:35.000Z (about 1 year ago)
- Default Branch: master
- Last Pushed: 2025-02-04T08:22:31.000Z (about 1 year ago)
- Last Synced: 2025-02-13T17:52:07.563Z (12 months ago)
- Topics: mechine-learing, numpy, pandas, python, scikit-learn, scipy, streamlit
- Language: Jupyter Notebook
- Homepage: https://spam-shield-dz46.onrender.com/
- Size: 299 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 🚀 **SpamShield: SMS Spam Detection**
🔍 **Predict Spam or Not in a Click!**
## 📌 **Overview**
SpamShield is an AI-powered SMS spam detection model that analyzes incoming messages and predicts whether they are **spam 🚨** or **safe ✅**. Built with **Python & Machine Learning**, this tool helps keep your inbox clean and spam-free!
🎯 **What It Does:**
- Takes an SMS as input 📩
- Predicts **Spam or Not Spam** instantly ⚡
- Uses a robust **Machine Learning model** for accuracy 🎯
- Hosted on the web using **Streamlit** 🌐
📢 **Try the Live Demo** 👉 [SpamShield](https://spam-shield-dz46.onrender.com/)
---
## 🛠️ **Tech Stack**
🔹 **Programming Language:** Python 🐍
🔹 **ML Libraries:** Scikit-learn 🤖, Pandas 📊, NumPy 🔢
🔹 **Web Deployment:** Streamlit 🚀
---
## ⚡ **Key Features**
✅ **Data Collection & Preprocessing** 🗂️
✅ **Exploratory Data Analysis (EDA)** 📊
✅ **Machine Learning Model Selection** 🤖
✅ **Web Interface for Predictions** 🌍
✅ **Deployed & Ready to Use!** 🎯
---
## 📊 **Data Collection & Processing**
📌 **Dataset Source:** [Kaggle SMS Spam Collection](https://www.kaggle.com/datasets/uciml/sms-spam-collection-dataset)
📌 **5,500+ labeled SMS messages** 📨
📌 **Preprocessing Steps:**
- Handling missing & duplicate values 🧹
- Label encoding the "type" column 🏷️
- Tokenizing text ✂️
- Removing **special characters, stopwords & punctuation** 🔍
- Converting text to lowercase 🔡
- Stemming words for better feature extraction 🌱
---
## 📈 **Exploratory Data Analysis (EDA)**
We analyzed the dataset using various techniques:
📊 **Visualizations:** Bar Charts 📊 | Pie Charts 🥧 | Heatmaps 🔥
📜 **Word Frequency Analysis** 📝
🌎 **Word Cloud for Spam & Non-Spam Messages** ☁️
📌 **Message Length Analysis (Characters, Words, Sentences)**
---
## 🏆 **Machine Learning Model Selection**
We tested multiple classifiers to find the best:
✅ **Naïve Bayes** 🟢
✅ **Random Forest** 🌳
✅ **K-Nearest Neighbors (KNN)** 📌
✅ **Decision Tree** 🌱
✅ **Logistic Regression** 📉
✅ **ExtraTreesClassifier** 🍃
✅ **Support Vector Classifier (SVC)** 💻
🔥 **Best Model:** Achieved **100% Precision** for spam detection!
---
## 🌐 **Web Deployment**
💡 **Built with Streamlit** for a smooth user experience!
📝 **How It Works:**
1️⃣ Enter an SMS message
2️⃣ Click **Predict**
3️⃣ Get instant results (**Spam 🚨 or Not Spam ✅**)
🔗 **Live Demo:** [SpamShield](https://spam-shield-dz46.onrender.com/)
---
## 🛠️ **Installation & Usage**
Want to run SpamShield locally? Follow these steps:
### 1️⃣ **Clone the Repository**
```
git clone https://github.com/Kane-dylan/SpamShield.git
cd SpamShield
```
### 2️⃣ **Install Dependencies**
```
pip install -r requirements.txt
```
### 3️⃣ **Run the App**
```
streamlit run app.py
```
💡 Now open your browser and start detecting spam! 🚀
---
## 🤝 **Contributions & Feedback**
I ❤️ contributions! If you:
- ✅ **Found a bug** 🐞
- ✅ **Have a feature request** 🌟
- ✅ **Want to improve the model** 📈
Feel free to open an issue or pull request at 👉 [GitHub Repo](https://github.com/Kane-dylan/SpamShield.git)
🚀 Let's build a spam-free future together!