🚀 Freasher in Data Science | Turning data into meaningful insights.
- 🌟 I am a fresher in Data Science, passionate about uncovering stories hidden in data.
- 🧑💻 Experienced with Python, Pandas, NumPy, and machine learning, Deep learning frameworks.
- 🔢 Love building data-driven solutions and interactive dashboards.
- 🌱 Currently expanding my skills in deep learning and cloud data platforms.
- 💬 Ask me about data wrangling, visualization, or finding insights!
- 📫 How to reach me: Email | LinkedIn
- ⚡ Fun fact: My favorite charts are violin plots!
A Machine Learning project that classifies whether a message is Spam or Not Spam.
Key Highlights:
-
✅ Built using Python, Scikit-learn, and NLP techniques.
-
📊 Preprocessed text data (stopwords removal, stemming, vectorization using TF-IDF).
-
🤖 Trained multiple models (Naive Bayes, Logistic Regression, etc.) to find the best performer.
-
⚡ Integrated into an interactive Streamlit web app for real-time message classification.
-
🌐 Deployed for easy access and testing.
A Deep Learning project that classifies images into Cat 🐱 or Dog 🐶 using Convolutional Neural Networks (CNN).
Key Highlights:
- 🧠 Built with TensorFlow/Keras and CNN architecture.
- 📷 Preprocessed and augmented image dataset for robust training.
- ⚡ Achieved high accuracy on validation & test data.
- 📊 Visualized training performance with accuracy/loss curves.
- 🚀 Can be extended into a real-time image classification web app.
A Data Science + Machine Learning project that predicts the winning probability of IPL teams during a live cricket match.
Key Highlights:
- 📊 Analyzed ball-by-ball IPL datasets to extract match insights.
- ⚡ Built a machine learning model to calculate real-time win probabilities.
- 🧮 Considered factors like runs, overs, wickets, current run rate, and required run rate.
- 🚀 Designed an interactive visualization dashboard for probability tracking.
- 🌐 Future-ready for deployment as a live match predictor app.
A Machine Learning + Deep Learning project to classify whether a breast tumor is Malignant (cancerous) or Benign (non-cancerous).
Key Highlights:
- 🧠 Used Logistic Regression for classification.
- 📊 Performed feature engineering on medical datasets (mean radius, texture, smoothness, etc.).
- ⚡ Achieved high accuracy, precision, and recall in detecting cancer.
- 📈 Compared multiple ML model to select the most reliable predictor.
- 🌐 Can be deployed as a Streamlit web app for real-time cancer prediction support.
