Projects
Unsupervised Learning Model This project focuses on analyzing Credit Card Dataset for Clustering. The workflow is divided into two main parts:
- Unsupervised Learning (K-Means Clustering)
- Used to segment customers based on their credit card usage patterns.
- Supervised Learning (Random Forest Classification)
- Built a classification model on the clustered dataset to predict cluster labels.
- Achieved 96.98% accuracy .
This project focuses on predicting the price range of mobile phones based on their technical specifications. The goal was to identify the most accurate machine learning model for this classification task by experimenting with multiple algorithms.
Methodology:
- Data preprocessing and feature selection.
- Trained and evaluated multiple ML models to compare performance.
- Selected the best-performing model for deployment.
At the end, the Logistic Regression Algorithm achieved the highest accuracy of 97.25%, proving to be the most effective model for predicting mobile phone price ranges.This project focuses on predicting the price range of mobile phones based on their technical specifications. The goal was to identify the most accurate machine learning model for this classification task by experimenting with multiple algorithms.
Methodology:
- Data preprocessing and feature selection.
- Trained and evaluated multiple ML models to compare performance.
- Selected the best-performing model for deployment.
At the end, the Logistic Regression Algorithm achieved the highest accuracy of 97.25%, proving to be the most effective model for predicting mobile phone price ranges.This project focuses on predicting the price range of mobile phones based on their technical specifications. The goal was to identify the most accurate machine learning model for this classification task by experimenting with multiple algorithms.
SafeHaven is a smart, AI-powered system built to safeguard elderly individuals through continuous monitoring and proactive alerts. It integrates a cross-platform mobile application that allows caregivers and family members to stay connected anytime, anywhere. The system includes a wearable smart watch that monitors vital signs (such as Heart Rate, Body Temperature, spo2) to predict falls before they happen, offering critical early intervention.
SafeHaven also features a smart belt with embedded sensors and a camera-based computer vision system that work together to accurately detect real-time falls and sudden health-related events. When a fall or anomaly is detected, the system sends instant notifications to caregivers, ensuring timely assistance.
By combining IoT, machine learning, and computer vision, SafeHaven delivers a complete and compassionate solution that empowers independent living while ensuring safety, reliability, and peace of mind.
SafeHaven also features a smart belt with embedded sensors and a camera-based computer vision system that work together to accurately detect real-time falls and sudden health-related events. When a fall or anomaly is detected, the system sends instant notifications to caregivers, ensuring timely assistance.
By combining IoT, machine learning, and computer vision, SafeHaven delivers a complete and compassionate solution that empowers independent living while ensuring safety, reliability, and peace of mind.
A desktop application that automates posting tweets to X (formerly Twitter) using ChatGPT to generate tweet content from user-provided input.
Key features include:
- Image attachment from a user-specified folder (ordered by date, ascending or descending)
- Customizable tweet count and intervals between posts
- Optional links within tweets
- History log of generated and posted content for reuse
developed a sustainable green hydrogen system powered by solar energy through water electrolysis, integrating IoT sensors to monitor vital parameters like temperature, humidity, and hydrogen output. The system uses AI and machine learning to predict weather conditions and optimise energy use, ensuring high efficiency and minimal environmental impact.
A cross-platform password generator app that lets users create strong, customizable passwords by selecting length and character types (uppercase, lowercase, numbers, special characters). It evaluates password strength, saves passwords locally for privacy, and features an offline-friendly, user-friendly interface compatible with Windows, macOS, Linux, Android, and iOS.
This project utilises a dataset containing 8 key features, including age, glucose level, pulse rate, gender, troponin levels, KCM, high blood pressure, and low blood pressure to predict the likelihood of a heart attack. The model is designed to analyse these medical indicators and determine the patient's health status. It achieved an impressive accuracy of 96.3%, delivering reliable results during testing and validation.
Achievements
6th Olympiad (ITA - SDGs 2024) in Damietta
the Air Defense Competition(ISEIC 2024) in Cairo
ITC 2025 in New Administrative Capital
DIGITOPIA First version 2025
Testimonials
Get in Touch
I am always eager to connect, discuss new opportunities, and explore how my expertise in cross-platform development and Artificial Intelligence can help bring your vision to life. Reach out through any of the channels below.
