Munalisa Paul

Data Science Projects 

PRCP-1016 Heart Disease PredictionPRCP-1016 Heart Disease Prediction

• Leveraged advanced machine learning algorithms to develop a predictive model that accurately identifies the likelihood of heart disease based on patient data. Skills utilized include data analysis, predictive modeling, and proficiency in Python.

PRCP-1020 House Price PredictionPRCP-1020 House Price Prediction

• Built a sophisticated regression model to predict house prices based on multiple variables such as location, size, and amenities. My expertise in statistical analysis, model tuning, and data visualization were key to the success of this project.

PRCP-1025 Flight Price PredictionPRCP-1025 Flight Price Prediction

• Created an AI-driven tool for predicting flight prices in real-time, helping consumers make cost-effective travel decisions. This project enhanced my capabilities in data scraping, machine learning, and algorithm optimization.

PRCP-1027 Skin Disorder IdentificationPRCP-1027 Skin Disorder Identification

• Developed an image recognition system to diagnose skin disorders using deep learning techniques. This work showcased my skills in computer vision, neural networks, and the application of AI in healthcare diagnostics.

Masters Project

1.Deep learning based-police bladed weapon analysis system (June 2024 - September 2024)

• Constructed a novel dataset of images from various sources to form the foundation of a knife identification system
• Implemented and evaluated state-of-the-art image retrieval systems on the specialized dataset
• Developed and validated a novel deep learning architecture based on convocational neural networks that significantly improved image retrieval accuracy
• Created a novel architecture that effectively extracts features from vision transformers and integrates advanced aggregation methods to enhance overall system performance

Bachelors Projects

2.AIMLweyShield - Artificial Intelligence Based System for Weapon Detection and Identification for Military Applications (Jan 2023 - May 2023)

• This project is a system based on artificial intelligence and machine learning for Weapon Detection and Identification for Military Applications is designed as a surveillance system to detect four various types of weapons, namely AK 47, Insas Rifle, Rocket Launcher and Grenade.
• The datasets used to make this artificial intelligence Based framework incorporate promptly accessible dataset from Kaggle and modified dataset from imdfb. The Consequences be damned V6 model is utilized to make this framework for adaptable and productive item identification and grouping. The dataset for new and raw data can be taken from “YawCam” that will capture the picture of the area or place where this project is implemented every after 10 seconds also this will record the motion of the object.
• The user can use the system to detect and identify the selected weapons by uploading an image or MP4 format video into the system. It can also help to detect weapons in real-time, with one click on the button, it will open the camera automatically.

3.Contactless Temperature monitoring system (Dec 2021 - Apr 2022)

• Developed an Arduino Uno and Python-based contactless temperature monitoring system using OpenCV. Data is recorded in a MySQL database.

4. Arduino Uno based Parking Sensor (Sep 2021 - Feb 2022)

• Arduino Uno based parking sensor with a database of how many cars are entering a certain parking area.

5.Arduino Uno based Watch (Jan 2020 - Mar 2020)

• Arduino Uno and Python-based watch which had functionalities of a clock, alarm, stopwatch and timer.