Portfolio - Neema Rose

Hello, I am

Neema Rose

Data Science/AI Enthusiast

About Me.

I'm a freelance data analyst with a Master's degree in Statistics, passionate about transforming complex data into actionable insights.
Currently, I'm expanding my expertise into Generative AI and Data Science, exploring how cutting-edge technologies can solve real-world problems. I'm continuously building my skills in machine learning, AI applications, and advanced analytics to stay at the forefront of this rapidly evolving field.
I thrive on tackling challenging datasets, uncovering hidden patterns, and communicating findings in ways that stakeholders can understand and act upon. Whether it's through predictive modeling, dashboard development, or exploratory analysis, I'm committed to delivering insights that matter.

2

Machine Learning Projects

1

Deep Learning Project

Skills.

Python

SQL

R Programming

Excel

Power BI

Statistical Analysis

Hypothesis Testing

Machine Learning

Deep learning

IBM SPSS Statistics

Projects.

My Experience.

📊Data Analyst | Freelance — Remote

Apr 2023 – Present

Delivering data-driven insights and statistical analysis for clients across healthcare, finance, and retail sectors, specializing in research consulting and business intelligence solutions.

Key Technologies: Python, R, SPSS, Excel, Pandas, NumPy, Matplotlib, Seaborn, Statistical Analysis

💻Machine Learning Engineer Intern | Smartinternz — Remote

Nov 2025 – Dec 2025

Developed and deployed end-to-end machine learning solutions for predictive analytics, focusing on classification algorithms and web-based model deployment.

Key Technologies: Python, Scikit-learn, Pandas, Flask, Git/GitHub, Machine Learning Algorithms, Model Evaluation, Web Development

🏛️Data Analyst Intern | Local Self Government Department Joint Director Office — Thrissur, Kerala

Apr 2024 – May 2024

Conducted a comprehensive survey analysis for the government waste management program, providing statistical insights and policy recommendations to improve operational efficiency.

Key Technologies: Statistical Analysis, Survey Design, Data Collection, SPSS/Excel, Hypothesis Testing, Report Writing

Awards & Certificate.

Python: Beginner to Advanced For Data Professionals

SQL for Data Science

Math and Statistics For AI, Data Science

Master Machine Learning for Data Science & AI: Beginner to Advanced

Deep Learning: Beginner to Advanced

Natural Language Processing

Gen AI to Agentic AI with Business Projects

Contact.

Feel free to get in touch with me. I am always open to discussing new projects, creative ideas or opportunities to be part of your visions.

Download CV

[email protected]

https://linkedin.com/in/neemarose

GitHub

https://github.com/neema-rose

Machine Learning Project

Health Insurance Premium Prediction System

Domain/Function: Insurance Analytics

Built an end-to-end machine learning system to predict health insurance premiums using customer health and demographic data. Implemented age-based model segmentation and deployed the solution as an interactive Streamlit web application for real-time underwriting support.

View Project
on GitHUB

View Project
on

Health Insurance Premium Prediction System

Interactive Streamlit-based web application that allows underwriters to input customer details such as age, BMI, income, smoking status, and medical history to generate health insurance premium predictions.

Real-time premium prediction generated by the machine learning model based on user inputs, demonstrating the system’s accuracy, responsiveness, and practical use for insurance underwriting decisions.

Project Details

Developed an end-to-end machine learning system to predict health insurance premiums for Shield Insurance. The project followed a complete ML lifecycle, starting from data loading, cleaning, exploratory data analysis, and feature engineering. A key innovation was age-based segmentation, where separate models were trained for customers below 25 and above 25 years to capture different health risk patterns. Multiple regression models were trained and fine-tuned, achieving approximately 98% accuracy with minimal extreme prediction errors after detailed error analysis. The final models were serialized and integrated into an interactive Streamlit web application that allows underwriters to input customer details such as age, BMI, smoking habits, income, and medical history to receive real-time premium estimates. The solution delivers a production-ready MVP and forms a scalable foundation for future straight-through processing (STP) automation.

Deep Learning Project

Car Damage Detection using Deep Learning

Domain/Function: Computer Vision (Automotive AI)

AI-powered computer vision system that analyzes car images to detect and classify front and rear vehicle damage using a deep learning model deployed as a real-time web application

View Project
on GitHUB

View Project
on

Car Damage Detection using Deep Learning

AI-powered vehicle damage classifier that predicts front Breakage using Deep Learning and ResNet-50, with a Streamlit web UI for real-time image predictions.

AI-powered vehicle damage classifier that predicts rear damage type (Normal, Breakage, Crushed) using Deep Learning and ResNet-50, with a Streamlit web UI for real-time image predictions.

Project Details

This project is an AI-powered vehicle damage detection system that classifies car images into six categories: Front Normal, Front Breakage, Front Crushed, Rear Normal, Rear Breakage, and Rear Crushed. Built using deep learning and transfer learning with ResNet-50, the model learns visual patterns of dents, cracks, and crushed vehicle parts from labeled car images. The trained model is deployed through a Streamlit web application where users can upload an image and instantly receive a damage prediction. This Proof of Concept was designed for automotive inspection use cases, demonstrating how computer vision can automate manual vehicle assessment, reduce human error, and improve inspection speed in car resale and insurance workflows.

Machine Learning Project

Predictive Pulse: Harnessing Machine Learning for Blood Pressure Analysis

Domain/Function: Healthcare Analytics

A machine learning–based healthcare application that predicts a person’s risk of hypertension using clinical and lifestyle data, with results displayed via a web interface for easy access.

View Project
on GitHUB

View Project
on

Predictive Pulse: Harnessing Machine Learning for Blood Pressure Analysis

Web interface of the Hypertension Prediction system where users enter patient health parameters such as age, BMI, glucose level, and lifestyle factors to generate a hypertension risk prediction.

Prediction result page displaying the model’s output indicating whether the patient is at risk of hypertension, along with the calculated health-based classification

Project Details

This project focuses on predicting the likelihood of hypertension using machine learning techniques applied to patient health data. The system analyzes key clinical attributes such as age, BMI, glucose level, smoking status, physical activity, and blood pressure indicators to classify whether a person is at risk of developing hypertension. The objective is to support early detection and preventive healthcare by providing a fast and data-driven risk assessment.

The model was trained using supervised learning algorithms and optimized to achieve high prediction accuracy. The trained model was deployed using a Flask-based web application, allowing users to input health parameters and receive real-time predictions. This end-to-end solution demonstrates the complete machine learning lifecycle, including data preprocessing, model training, evaluation, and web-based deployment for practical use in healthcare analytics.