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.
| π Project | π Tech Stack | π 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 β Live Demo
- Diabetes Prediction β Live Demo
- Movie Recommendation System - Live Demo
- Gold Price Prediction - Live Demo
- caesar-cipher-app - Live Demo
- π§ Advanced Machine Learning Algorithms
- π Deep Learning
- π Cloud Deployment strategies (AWS, Render)
Authors: Shibdas Dutta, Subhrendu Guha Neogi, Diya Chanda, Arpan Pramanik, ΓzgΓΌn Girgin, Enes Ladin ΓncΓΌl
π DOI: https://doi.org/10.1109/ICRITO66076.2025.11241706
π§ Developed a multi-task deep learning framework for fruit classification and quality grading using a multi-headed CNN architecture.
π Achieved 98% classification accuracy and 99% quality detection accuracy.
π Integrated Grad-CAM for explainable AI and model interpretability.
π Deployed using Streamlit for real-time usability.
π 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
π DOI: https://doi.org/10.1109/ICRITO66076.2025.11241535
π± Designed a lightweight deep learning model for detecting quality in crops like potato and tomato.
π Achieved 99.9% crop classification accuracy and 98.5% quality detection accuracy.
π Utilized Grad-CAM visualizations for transparency and trust in predictions.
β‘ Built for real-time deployment using Streamlit in precision agriculture applications.
Authors: Shibdas Dutta, Barshan Adhikari, Arpan Pramanik, Diya Chanda
π DOI: https://doi.org/10.1109/COMPUTINGCON64838.2025.11376762
π§ Proposed a hybrid CNN-ViT architecture for simultaneous crop classification and quality assessment.
β‘ Achieved 30%+ parameter reduction, making the model efficient and lightweight.
π Reached 98.45% accuracy for potato and 97.49% for tomato classification, with 98.5% quality assessment accuracy.
π Focused on explainable AI techniques to improve transparency and trust in predictions.
"Machine learning is not magic; it's just math." π
Thanks for visiting my profile! Feel free to explore my repositories and connect with me. π