Proma Hossain Progga

Proma Hossain Progga

Hi! I am working as a Machine Learning Engineer at SSCL. I completed my undergraduate studies specializing in Data Science, graduating with the top position in Computer Science and Engineering at UIU. My current research focuses on computer vision and language. Feel free to contact me if you are interested in my work.

šŸ“– Educations

  • 2019.07 - 2023.05Ā  B.Sc. in Computer Science and Engineering, United International University
  • 2018Ā  Higher Secondary Certificate (Science), Gurrudayal Govt. College
  • 2016Ā  Secondary School Certificate (Science), SV Govt. Girls High School

šŸ“ Publications

Neurocomputing
sym

A bidirectional Siamese recurrent neural network for accurate gait recognition using body landmarks
In Neurocomputing Journal, 2023.
Proma Hossain Progga, Md. Jobayer Rahman, Swapnil Biswas, Md. Shakil Ahmed, Arif Reza Anwary and Swakkhar Shatabda.

[Paper]Ā  [Code]

ICONIP'22
sym

iResSENet: An Accurate Convolutional Neural Network for Retinal Blood Vessel Segmentation
In International Conference on Neural Information Processing, pp. 567-578. Cham: Springer International Publishing.
Proma Hossain Progga & Swakkhar Shatabda

[Paper]Ā  [Code]

ICCT'23
sym

K-Nearest Neighbour Classifier for Big Data Mining based on Informative Instances
In International Conference for Convergence in Technology (I2CT), pp. 1-7. IEEE.
Proma Hossain Progga, Md. Jobayer Rahman, Swapnil Biswas, Md. Shakil Ahmed and Dewan Md Farid

[Paper]Ā 

ICISS'23
sym

UAVs in Green Health Care for Energy Efficiency and Real-Time Data Transmission
In International Conference on Intelligent Sustainable Systems, pp. 773-788. Singapore: Springer Nature Singapore
Anika Khaer, Md Siam Hossain Sarker, Proma Hossain Progga, Saniyat Mushrat Lamim, and Md Motaharul Islam

[Paper]Ā 

šŸŽ– Honors and Awards

  • Summa Cum Laude - United International University (2024)
  • 100% Scholarship - United International University (2019-2023)
    • Awarded 100% scholarships for excellent academic performance in 10 out of 12 trimesters, which are given to the top 2% of students in specific programs.
  • 1st place in Asia, 13th Globally - University Rover Challenge (2022)
    • It is an international robotics competition where teams design and build Mars rovers to complete tasks related to exploration and research in simulated Martian environments.
  • 1st Runner-Up - CSE Project Showcase, United International University (Spring 2022)
    • Led and contributed significantly to the successful execution of the ā€œProtibaadā€ social awareness project.
  • 5th position - Bengali Automatic Speech Recognition (Kaggle Competition), DL Sprint - BUET CSE Fest (2022)

šŸ’» Academic Services

  • Part-Time Lecturer - United International University (June 2023 - September 2023)
    • CSE 1115: Object Oriented Programming. (Summerā€™23)
    • CSE 1325: Digital Logic Design. (Summerā€™23)
  • Undergraduate Teaching Assistant - United International University (March 2022 - May 2023)
    • CSE 2216: Data Structure and Algorithms I Laboratory. (Springā€™23)
    • CSE 3421: Software Engineering. (Fallā€™22)
    • CSE 3522: Database Management Systems Laboratory. (Fallā€™22)
    • CSE 2216: Data Structure and Algorithms I Laboratory. (Summerā€™22)
    • CSE 3522: Database Management Systems Laboratory. (Springā€™22)
  • Grader - United International University (November 2022 - May 2023)
    • CSE 4889: Machine Learning. (Springā€™23)
    • CSE 3811: Artificial Intelligence. (Springā€™23)
    • CSE 1111: Structured Programming. Language. (Springā€™23)
    • MATH 2205: Probability and Statistics. (Fallā€™22)

šŸ’¬ Selected Projects

  • Information Extraction From Financial Documents - Industry Project
    This project aims to extract entities from financial trade-related documents and organize the extracted information into a structured JSON object. This JSON object can then be utilized across various applications.
    • Domain: Classification, OCR, Segmentation, Detection, LLM, Multimodal
  • Vehicles Number Plate Detection and Recognition - Industry Project
    This project aims to extract entities from financial trade-related documents and organize the extracted information into a structured JSON object. This JSON object can then be utilized across various applications.
    • Domain: Detection, Multimodal, OCR
  • Retinal Blood Vessel Segmentation - University project
    The primary objective of this research is to identify a segmentation methodology that surpasses existing methods in terms of performance on multiple standard benchmark datasets.
    • Language: Python, Keras (Tensorflow as backend)
    • Neural Networks: U-Net, Squeeze-and-Excitation, Deep Residual Network, Inception Network, ConvNeXt, and so on.
    • Dataset : DRIVE, CHASEDB1, STARE, and HRF
  • Graph Neural Network to Analyze Protein Secondary Structure - University project
    This research used GNNs to predict protein secondary structure and compared several models.
    • Language and Frameworks: Python, Tensorflow.
    • Algorithms: Graph Neural Network, Support Vector Machine, AdaBoost, K-Nearest Neighbours, Decision Tree.
    • Dataset: ccPDB 2.0
  • Bengali Automatic Speech Recognition - Kaggle Competition
    Built a model to recognize Bengali speech from out-of-distribution audio using a crowdsourced dataset with 400 hours of speech. The project aimed to enhance Bengali speech recognition and address out-of-distribution generalization challenges.
    • Framework: PyTorch
    • Word Embedding & Language model: BERT, Word2Vec, N-Gram (6-gram)
    • Dataset: Bengali Common Voice Speech
  • Biological Life Detection through Rock Classification - University Rover Challenge
    • The science team developed a subsystem to categorize soil and rock samples into extant, extinct, or ā€˜No Presence of Lifeā€™. Our research utilized machine learning (ML) and deep learning (DL) methodologies and we also developed our own dataset.
  • Restaurant Management System - University project
    This system allows customers to reserve tables at various restaurants based on availability. Customers can order from the restaurantā€™s online menu and either serve themselves or request table service for an additional fee. The system generates bills and allows for online payment. Admins can manage restaurants, menus, and orders, and generate reports.
    • Backend Language: Python (Django).
    • Frontend: HTML, Sass, CSS, JavaScript, jQuery
  • Protibaad - University project
    Protibaad is a comprehensive software system that enhances public safety and awareness. It features a crime alert system, missing person reporting, a survey pool, and a blog section. With its user-friendly interface and range of useful features, Protibaad is an invaluable tool for promoting safety, community engagement, and social awareness.
    • Backend Language: PHP, Ajax, JSON.
    • Frontend: HTML, CSS, Javascript.
    • Database: SQL.
  • E-commerce Desktop Application - University project
    It is a desktop application where multiple customers can purchase and make payments at a time like e-commerce.
    • Library: Java FX
    • Communication Interface: Socket.
  • University Management System - University project
    This Project is based on Java to understand the core concept of object-oriented programming. Features: ā€¢ Faculty Management (add, remove, or show). ā€¢ Student Management (Insert, remove, or show). ā€¢ Course Transactions (add, drop, or show
    • Language: Java