https://github.com/raju-2003/indiaai-cyberguard-ai-hackathon

An NLP-powered system to simplify cybercrime reporting by analyzing descriptions, categorizing incidents, and providing actionable insights.
https://github.com/raju-2003/indiaai-cyberguard-ai-hackathon

matplotlib nltk numpy pandas python random-forest-classifier re scikit-learn seaborn shap spacy wordcloud

Last synced: about 1 month ago
JSON representation

An NLP-powered system to simplify cybercrime reporting by analyzing descriptions, categorizing incidents, and providing actionable insights.

• Host: GitHub
• URL: https://github.com/raju-2003/indiaai-cyberguard-ai-hackathon
• Owner: raju-2003
• Created: 2024-11-22T15:53:58.000Z (about 1 month ago)
• Default Branch: main
• Last Pushed: 2024-11-22T16:10:39.000Z (about 1 month ago)
• Last Synced: 2024-11-22T16:37:50.689Z (about 1 month ago)
• Topics: matplotlib, nltk, numpy, pandas, python, random-forest-classifier, re, scikit-learn, seaborn, shap, spacy, wordcloud
• Language: Jupyter Notebook
• Homepage: https://drive.google.com/file/d/1mwxB3uQAXCGdjd3nYwsymlLlXZ4fcv8y/view?usp=sharing
• Size: 0 Bytes
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# IndiaAI CyberGuard AI Hackathon

# Cybercrime Reporting NLP System

This project leverages advanced **Natural Language Processing (NLP)** techniques to enhance the process of filing cybercrime reports on the **National Cyber Crime Reporting Portal (NCRP)**. The system streamlines the reporting process by analyzing textual descriptions and supporting media files, identifying themes, and ensuring accurate and user-friendly submissions.

---

## 📌 Project Goals

- Simplify the process of filing cybercrime reports.

- Accurately classify cybercrime categories based on user descriptions.

- Provide actionable insights to law enforcement using data trends.

---

## 🚀 Features

- **Sentiment Analysis**: Detects tone and urgency of reports to prioritize cases.

- **Topic Modeling**: Identifies commonly reported themes like phishing, identity theft, and financial fraud.

- **Text Classification**: Categorizes criminal activities with high accuracy using a Random Forest Classifier.

- **Visual Insights**: Generates word clouds, confusion matrices, and sentiment trend graphs.

---

## 🛠️ Tech Stack

- **Programming Language**: Python

- **Libraries**: 

  - Data Handling: `Pandas`, `NumPy`

  - Text Processing: `spaCy`, `NLTK`, `re`

  - Machine Learning: `Scikit-learn`, `TF-IDF Vectorizer`

  - Visualization: `Matplotlib`, `Seaborn`, `WordCloud`

---

## 🔄 Workflow

1. **Data Loading and Cleaning**  

   - Load datasets, remove duplicates, and handle missing values.

2. **Exploratory Data Analysis (EDA)**  

   - Generate word clouds and analyze sentiment trends.

3. **Text Preprocessing**  

   - Perform tokenization, lemmatization, stopword removal, and feature extraction.

4. **Model Training and Evaluation**  

   - Train the Random Forest Classifier and evaluate using precision, recall, and F1-score.

5. **Visualization**  

   - Display feature importance, confusion matrices, and sentiment trends.

---

## 📊 Results and Insights

- The system captures key cybercrime themes and provides structured feedback for users.

- Sentiment analysis identifies peaks in negative sentiment to indicate spikes in criminal activity.

- The Random Forest Classifier achieves robust accuracy and handles high-dimensional text data effectively.

---

## 💡 How to Use

1. Download and Use:

   ```bash

   https://drive.google.com/file/d/1mwxB3uQAXCGdjd3nYwsymlLlXZ4fcv8y/view?usp=sharing