🏥 IoT-Based Patient Monitoring System for Healthcare (09/2020 - 02/2021)
- Developed a real-time IoT-based patient monitoring system to track vital health parameters such as heart rate, body temperature, and oxygen saturation levels.
- Integrated biomedical sensors (heart rate sensor, temperature sensor, pulse oximeter) with a microcontroller (Arduino/Raspberry Pi) for continuous data collection.
- Transmitted sensor data wirelessly to a cloud platform using Wi-Fi and MQTT protocols, enabling remote monitoring for healthcare professionals.
- Designed and implemented a user-friendly web dashboard for real-time data visualization using Firebase and ThingSpeak.
- Implemented an alert system that triggers SMS or email notifications in case of abnormal health conditions.
- Ensured system efficiency and reliability through rigorous testing and calibration of sensors.
Technologies Used: Arduino, Raspberry Pi, IoT Sensors, Python, Firebase, ThingSpeak, MQTT, HTML/CSS, JavaScript.
📊 Exploratory Data Analysis of Hotel Booking Trends (08/2022 - 09/2022)
- Conducted a comprehensive exploratory data analysis (EDA) on hotel booking datasets to understand booking patterns, customer preferences, and factors influencing cancellations.
- Preprocessed data by handling missing values, encoding categorical variables, and normalizing numerical features.
- Analyzed key trends such as lead time, seasonality, market segments, and the impact of special requests on cancellations.
- Created insightful visualizations using Matplotlib and Seaborn, showcasing trends like booking cancellations, customer demographics, and revenue impact.
- Derived business insights to help hotel managers improve occupancy rates, optimize pricing strategies, and reduce cancellation rates.
Technologies Used: Python, Pandas, NumPy, Matplotlib, Seaborn, Jupyter Notebook.
🔗 View Code
📱 Predicting Mobile Price Ranges with Machine Learning Models (10/2022 - 12/2022)
- Designed a machine learning model to predict mobile phone price ranges based on technical specifications such as battery power, RAM, processor speed, and camera quality.
- Performed data preprocessing, including feature selection, outlier detection, and standardization to improve model performance.
- Trained and evaluated multiple classification models, including Decision Trees, Random Forest, Support Vector Machine (SVM), and Gradient Boosting.
- Optimized model performance using hyperparameter tuning and cross-validation techniques.
- Deployed the model using Flask, allowing users to input specifications and receive real-time price predictions.
Technologies Used: Python, Scikit-Learn, Pandas, NumPy, Matplotlib, Flask, Hyperparameter Tuning, Classification Models.
🔗 View Code
🚲 Bike Sharing Demand Forecasting Using Machine Learning (01/2023 - 03/2023)
- Developed a predictive model to forecast daily bike rental demand based on historical usage data, weather conditions, and temporal features.
- Performed extensive data cleaning, handling missing values, feature engineering, and transforming categorical variables.
- Conducted time-series analysis and correlation studies to identify the impact of external factors (temperature, humidity, season) on bike demand.
- Implemented regression models including Linear Regression, Decision Trees, Random Forest, and XGBoost to predict demand.
- Applied hyperparameter tuning and cross-validation techniques to enhance model accuracy.
- Created interactive visualizations to analyze trends and forecast demand, aiding operational decision-making for bike-sharing companies.
Technologies Used: Python, Scikit-Learn, Pandas, NumPy, XGBoost, Time-Series Analysis, Matplotlib, Seaborn.
🔗 View Code
🛒 Customer Segmentation for Online Retail Using Machine Learning (03/2023 - 05/2023)
- Conducted customer segmentation for an online retail dataset using unsupervised learning techniques to identify distinct customer groups based on purchasing behavior.
- Preprocessed data by handling missing values, scaling numerical features, and encoding categorical data.
- Applied clustering techniques such as K-Means Clustering, Hierarchical Clustering, and DBSCAN to segment customers effectively.
- Used the Elbow Method and Silhouette Score to determine the optimal number of clusters.
- Generated insights on customer spending patterns, product preferences, and frequency of purchases, aiding in personalized marketing strategies.
- Designed a dashboard with interactive visualizations to present segmentation results and recommendations for business decision-making.
Technologies Used: Python, Scikit-Learn, Pandas, NumPy, K-Means Clustering, Hierarchical Clustering, DBSCAN, Matplotlib, Seaborn.
🔗 View Code