Welcome to my GitHub profile!
I'm Diya Chanda, an aspiring AI/ML Engineer, passionate about crafting intelligent solutions and building impactful projects. Currently pursuing B.Tech in CSE (AIML) at The Neotia University, I enjoy working on Machine Learning and Deep Learning, and I'm always eager to learn new technologies.
| 🔍 Project | 🚀 Tech Stack | 🔗 Repository |
|---|---|---|
| Car Price Prediction | Streamlit, Python, Scikit-learn | GitHub Repo |
| Diabetes Prediction | Streamlit, Python, Scikit-learn | GitHub Repo |
| Movie Recommendation System | Streamlit, Python, Scikit-learn, Cosine-similarity | GitHub Repo |
| Attendance Tracker | Python, Tkinter, Pandas | GitHub Repo |
| MNIST Digit Classification | Python, Matplotlib, Seaborn, OpenCV, TensorFlow | GitHub Repo |
| Gold Price Prediction | Python, Sklearn, Streamlit | GitHub Repo |
| caesar-cipher-app | Python, pip, Streamlit | GitHub Repo |
Here are some live demos of my projects that you can explore:
- Car Price Prediction – Live Demo
- Diabetes Prediction – Live Demo
- Movie Recommendation System - Live Demo
- Gold Price Prediction - Live Demo
- caesar-cipher-app - Live Demo
- 🧠 Advanced Machine Learning Algorithms
- 🌐 Deep Learning
- 🚀 Cloud Deployment strategies (AWS, Render)
Authors: Shibdas Dutta, Subhrendu Guha Neogi, Diya Chanda, Arpan Pramanik, Özgün Girgin, Enes Ladin Öncül
🔗 DOI: https://doi.org/10.1109/ICRITO66076.2025.11241706
🧠 Developed a multi-task deep learning framework for fruit classification and quality grading using a multi-headed CNN architecture.
📊 Achieved 98% classification accuracy and 99% quality detection accuracy.
🔍 Integrated Grad-CAM for explainable AI and model interpretability.
🚀 Deployed using Streamlit for real-time usability.
📄 CropSense: Explainable Deep Learning Framework for Accurate Quality Detection in Solanaceous Crops
Authors: Shibdas Dutta, Subhrendu Guha Neogi, Shiladitya Chowdhury, Vikrant Chole, Arpan Pramanik, Diya Chanda
🔗 DOI: https://doi.org/10.1109/ICRITO66076.2025.11241535
🌱 Designed a lightweight deep learning model for detecting quality in crops like potato and tomato.
📊 Achieved 99.9% crop classification accuracy and 98.5% quality detection accuracy.
🔍 Utilized Grad-CAM visualizations for transparency and trust in predictions.
⚡ Built for real-time deployment using Streamlit in precision agriculture applications.
Authors: Shibdas Dutta, Barshan Adhikari, Arpan Pramanik, Diya Chanda
🔗 DOI: https://doi.org/10.1109/COMPUTINGCON64838.2025.11376762
🧠 Proposed a hybrid CNN-ViT architecture for simultaneous crop classification and quality assessment.
⚡ Achieved 30%+ parameter reduction, making the model efficient and lightweight.
📊 Reached 98.45% accuracy for potato and 97.49% for tomato classification, with 98.5% quality assessment accuracy.
🔍 Focused on explainable AI techniques to improve transparency and trust in predictions.
"Machine learning is not magic; it's just math." 😎
Thanks for visiting my profile! Feel free to explore my repositories and connect with me. 🚀