MS-in-Data-Science

This repository serves as a comprehensive collection of the coursework and projects completed during my master's program. It showcases the extensive body of work that I have developed, demonstrating my proficiency and expertise in various areas of study.

Note:-

• Please note that while this repository contains my personal contributions and implementations, it does not include any assignment descriptions or materials provided by the university. I have ensured that only my own work is shared, respecting the intellectual property and confidentiality of the coursework.

• By showcasing my coursework and projects, I aim to highlight my ability to apply theoretical concepts to practical scenarios, solve complex problems, and deliver high-quality software solutions. Each project within the repository represents a unique challenge that I have successfully tackled, employing industry-standard methodologies and best practices.

• Throughout my master's program, I have consistently demonstrated a strong commitment to continuous learning and professional growth. I have actively sought out opportunities to expand my knowledge and skills, staying up-to-date with the latest advancements in software engineering and related fields.

• I believe that this repository not only reflects my technical capabilities but also showcases my dedication to producing well-documented, thoroughly tested, and user-focused software. I have always prioritized delivering solutions that meet the needs of end-users, ensuring intuitive interfaces and seamless user experiences.

My academic Coursework and Projects

Data Science (DASC 5300)

• Acquired knowledge of Supervised and Unsupervised algorithms and a basics of Neural Network.

Assignements

1. Pandas Exercise and Manual implementation of Decision Trees by writing functions for Gini Impurity, Entropy and Information Gain. Repository
2. Exploratory Data Analysis and Classification using SVM. Repository

Foundations of Computing (DASC 5301)

• Learned Data structures and Algorithms as part of the coursework. The course included three data analysis projects each contains a Report, Presentation and a viva. Projects were group of 2 peoople.

Projects

1. NYC accident data analysis Repository
2. DBLP data analysis using Graphs Repository
3. IMDB data analysis using SQL Repository

Probability and Statistics (DASC 5302)

• Learned Descriptive Statistics, Discrete and Continouous Probability distributions, Hypothesis Testing and Hypothesis testing in Linear Regression.

Assignments:-

1. EXPERIMENTAL DATA COLLECTION AND DESCRIPTIVE STATISTICS PART - I
   • Collected real-world data, performed descriptive statistics and made assumptions about the population distribution from the results of visualizations and descriptive statistics.
2. EXPERIMENTAL DATA COLLECTION AND DESCRIPTIVE STATISTICS PART - II
   • Performed goodness of fit chi-square test on the sample data collected to determine population distribution. Comparisonal study with the results from Part Repository

Data science Project Managment (DASC 5303)

• Leared the workflow of a data science project, different managment processes, investment strategies.
• Data science project management project.

Machine Learning (CSE 6363)

• Studied about Supervised machine learning algorithms in depth.

Assignments

1. Linear Regression from scratch Implementation using only NumPy library Repository
2. Logistic Regression and Linear Discriminant Analysis implementation from scratch using NumPy library and plotting the decision boundaries using Mlxtend library. Repository
3. Multilayer Neural Network from scratch using numpy applying oops concepts. Repository
4. Decision Tree and Random Forest from scratch and Boosting ( Ada boost ) from scratch Repository
5. Project - Gym Exercise Recommendation App Repository
6. Text Summarization using BART. Repository

Neural Networks (CSE 5368)

Assignments

1. Multilayer Neural Network from scratch Implementation using only NumPy library
   • Calculated the derivates manually by using F(w+h) - f(w-h) Repository
2. Multilayer Neural Network from scratch Implementation using only TensorFlow and NumPy without using Keras Repository
3. Convolutional Neural Network using tensorflow and keras by adding layer by layer and making visualizations of output from manually written confusion matrix function. Repository
4. Convolutional Neural Network using Tensorflow and keras that supports layer addition and deletion. Repository
5. Deep Convolutional Generative Adversarial Network using Tensorflow Repository

Computer Vision (CSE 6363)

The course structure includes a total of 4 assignments and projects. My works are displayed below.

1. Canny edge detection from scratch
   • Created an algorithm from scratch to detect the edge and compared the results obtained from the OpenCV python package. This includes implementing fuction for each of the steps that are undertaken within the the canny edge detection before reaching the final results. Repository
2. Hough transform and corner detection
   • This includes hough transform to identify different shapes and the implementation of the harris corner detection. Repository
3. Image stitching and panorama creation
   • This assignment includes the process of creating a panorama by identifying the key points in different image and combining them to obtain the stitched image. Repository
4. 2D Face Reconstruction
   • This assignment includes the process of reconstructing a face with the help of different number of eigen vectors. This work compares the reconstruction ability of different combination of the eigen reconstructed faces. Repository
5. Narrating the Unseen: Real-Time Video Descriptions for Visually Impaired Individuals
   • This is the final project of the computer vision course. The paper includes the use of creation of the model setup where the visually impaired people can understand their surroundings in the best way compared to all other existing methods. The system uses GPT-4_vision to do the image captioning of the surrounding in real time. Repository