ASRA AHMED

ASRA AHMED

Class of 2021
BS Computer Science
Minor: Social Development and Policy

Aspiration Statement

I am looking for job opportunities in data analysis, integration, or engineering. I want to pursue my Masters in a field like social data science, which will allow me to further the knowledge that I have gained from both my major (computer science) and minor (Social Development and Policy).

Core Skills

  • Python and C++
  • Qualitative Research
  • Task Delegation and Management
  • Analytical Skills, Data Analytics and Database Management
  • Technical Documentation

Academic Awards / Achievements

  • Dean's List Fall 2020
  • Attended the University of Sussex International Summer School program on a 100% scholarship in 2019

Experience

Leadership / Meta-curricular

  • HU Gazette (2019-2020) - Editor-In-Chief
  • Omicron 2018 - Director of Publications
  • Yearbook Committee, 2019 - Member

Internship / Volunteer Work

  • Access Group - Software Development Intern
  • 10Pearls - Software Development Intern
  • Playground, Habib University - Content Writer
  • Habib University - Undergraduate Teaching Assistant

Publications / Creative Projects

  • Published a paper in Tezhib Undergraduate Research Journal that investigated the effects of polarization at Habib University
  • Conducted analysis on the state of education in Pakistan's rural areas using Python and Pandas. Visualized findings using Tableau
  • Developed a Pipelined RISC V Processor
  • Built a digital system for Baitussalam Basic Health Unit

Final Year Project

Project Title

Sat2Sat Image Conversion

Description

Our thesis addresses the data sparsity problem arising from a lack of freely and publicly available satellite data. Data that meets this criterion comes from a range of different satellites with different sensors and optics, spatial resolutions and revisit rates. Thus, typically, for machine learning and data science tasks, data from only a single model of satellites can be used. Our solution was to perform image-to-image translation between different satellite types, to create a larger dataset of images that can then be used in a variety of different applications. We collected data using Google Earth to construct a dataset that we are using to train a computer vision algorithm to interchangeably translate an image from one satellite type to another. This is done using the pix2pix algorithm, which uses a conditional Generative Adversarial Network (cGAN), and the U-NET architecture. Results will be displayed on a website, showing time-lapse videos for the individual satellites and the combined dataset from the predicted images.