Welcome to our AI and Machine Learning Repository!
🚀 Explore the World of AI/ML! This repository is a treasure trove of fascinating AI projects that blend innovation, creativity, and cutting-edge techniques—dive in and be inspired!
We're thrilled to have you here! 🌟
👩💼 Description:
"BAIDS is a behavior-based Android Intrusion Detection System designed to identify suspicious and malicious activities on Android devices using digital forensic techniques and machine learning.
The system analyzes artifacts extracted from Android logical images—such as app permissions, network activity (IPDR/DNS), SMS logs, call records, and GPS data—to detect anomalies indicative of malware, phishing attacks, SIM spoofing, or unauthorized access.
BAIDS correlates multi-source forensic evidence to build a unified behavioral profile of the device and applies anomaly detection models (e.g., Isolation Forest / Autoencoders) to flag deviations from normal usage patterns. A Streamlit-based dashboard enables investigators to upload evidence, visualize timelines, map GPS movements, view risk scores, and generate forensic-ready reports with hash verification for evidentiary integrity."
👾 Project Category: Machine Learning
🌟 Details:
- Android Intrusion Detection System
- Android logical images—such as app permissions, network activity (IPDR/DNS) etc
- Marketing, Administration and R&D Cost, Location data given
🏏 Description:
This project focuses on detecting malicious Android applications through static forensic analysis and machine learning.
The system extracts APK files from a logical Android image or folder and analyzes features such as requested permissions, API usage patterns, file entropy, cryptographic hashes, and PE/APK metadata.
Suspicious applications are identified using trained ML classifiers (e.g., SVM / Random Forest) and heuristic rules such as high-entropy payloads and use of dangerous APIs (e.g., SMS sending, device ID access). Optional threat intelligence integration allows hash comparison with known malware databases. Results are presented in an interactive GUI that highlights high-risk apps and supports exportable forensic reports.
👾 Project Category: Machine Learning
🌟 Details:
- APK files from a logical Android image or folder
- static forensic analysis
- ML
📚 Description:
This project aims to detect and investigate phishing SMS messages using natural language processing and digital forensics.
The system uses an NLP model (BERT-based / Hinglish-trained classifier) to classify SMS messages as phishing or legitimate, with confidence scoring and explainability.
Beyond classification, the tool performs forensic correlation by linking phishing SMS with related call logs, app launches, browser activity, and domain WHOIS data. Timeline-based visualization helps investigators reconstruct attack sequences and assess user impact. The system supports SMS extraction from logical images or live device input and generates court-ready analytical summaries.
👾 Project Category: Machine Learning
🌟 Details:
- NLP model (BERT-based / Hinglish-trained classifier)
- SMS extraction from logical images
- Live device input
🥹 Description:
A modern portfolio analysis platform built with React and Flask that turns raw market data into actionable financial intelligence.
It fetches real historical stock prices, aligns multi-asset time series, and computes professional-grade metrics like expected return, volatility, and Sharpe ratio, visualized through clean charts and dashboards.
Designed as a foundation for advanced quant features, QuantWise showcases strong engineering, data science, and finance fundamentals while reflecting a scalable, product-oriented architecture suitable for real-world financial tools.
👾 Project Category: Machine Learning
🌟 Details:
- React and Flask
- Finance
💳 Description:
A real-time computer vision photobooth built with React and Flask that automatically captures photos when users smile.
It streams webcam frames to a backend powered by MediaPipe face landmarks and a custom smile-detection heuristic, triggers intelligent countdowns,
captures multiple shots, and generates a downloadable photo strip with captions. The project highlights expertise in real-time systems, CV pipelines, frontend-backend coordination, and user-centric design—combining technical depth with a fun, polished product experience.
👾 Project Category: Machine Learning
🌟 Details:
- real-time computer vision photobooth
- Powered by MediaPipe face landmarkss
🏢 Description:
A production-ready data analytics web app that transforms raw WhatsApp chat exports into meaningful behavioral and social insights.
Built with Python, Pandas, and Streamlit, it parses messy real-world chat logs into structured data and delivers rich visual analytics including activity timelines,
participation heatmaps, user engagement rankings, word and emoji analysis, and Hinglish-aware text processing.
The project demonstrates strong skills in data cleaning, NLP preprocessing, visualization, and product thinking by turning unstructured text into an interactive analytics experience.
👾 Project Category: Machine Learning
🌟 Details:
- Built with Python, Pandas, and Streamlit
- NLP preprocessing
-
Fork this Repository
Click the Fork button to create your copy of this repository. -
Clone the Repository
git clone https://github.com/GDG-IGDTUW/AI-ML-2.git cd repo-name -
Install Dependencies
Navigate to the project folder you're interested in.
For example:cd Sentiment-AnalysisLoad the dataset and Install necessary Libraries
-
Make Your Contributions
- Perform EDA.
- Train models.
- Enhance Accuracy.
- Add features.
- Test your changes.
-
Submit a Pull Request
Push your changes and create a pull request to propose your contributions! 🎉
We ❤️ contributions! Follow these simple steps to contribute:
-
Browse through Issues and Choose any
Browse the Issues tab and comment on the one you'd like to work on. -
Clone the Repo, Make changes and Branch Out
Create a new branch for your changes:git checkout -b feature-name
-
Commit Your Work
Write clear and concise commit messages:git commit -m "Add: Feature description" -
Push and PR
Push your branch and create a pull request for review.
🌟 Tips for Contributors
- Follow the repository’s code style and structure.
- Keep ML model training scripts well-indented and include comments.
- Share any interesting results or insights in the pull request description.
- If you want an issue to be assigned to you, Tag us and mention so under the issue.
- Please be patient and Feel free to Tag the maintainer or collaborators for any queries. ❤️
Happy Coding and Collaborating!🚀❤️