Personal Profile

A fast learner, adaptable and reliable person with

excellent time management skills. I am self

motivated, well organized and a very hardworking

person who's seeking for a role in the field of data science, where I can utilize my skills in a

competitive environment and be able to grow.

Rakebun

Islam

Lucky

COMPUTER SCIENCE

GRADUATE

Skills

Programming Language - Python, Java, JavaScript

Machine Learning and Deep Learning

Data analysis, visualization and cleaning

Neural Networks and Natural Language Processing

Matplotlib, Pytorch, Tensorflow, Keras, Scikit Learn Databases - MySQL, NoSQL

Tools - Jupyter notebook, VS code, Colab, Git-Hub

Effective verbal and written skills

Contact Details

Phone: 880 017********

Narayanganj, Bangladesh

https://github.com/Rakebun-Lucky

www.linkedin.com/in/rakebun-lucky

Bachelor of Science in Computer Science

CGPA - 3.54

BRAC UNIVERSITY

Academic Background

ESTIMATING FLOOD SUSCEPTIBILITY OF BANGLADESH IN THE FUTURE YEAR USING MACHINE LEARNING

This is a research paper which predicts the water level of a particular area of Bangladesh. It predicts the next months possible water level as well as the flooding probability based on the current month's data. The research was done based on real world data. Thesis and Research Paper

Certifications

-Peer-to-Peer Protocols and Local Area Networks - (Coursera Certification)

-Neural Networks and Deep Learning - (Coursera Certification)

-Machine Learning Real World projects in Python - (Udemy Certification)

-Natural Language Processing Real-World Projects in Python - (Udemy Certification)

-Programming for Everybody (Getting started with Python) - (Coursera Certification)

-Python Data Structures - (Coursera Certification)

-Fundamentals of Network Communication - (Coursera Certification) PREDICT THE STATUS OF CHRONIC KIDNEY DISEASE

This machine learning model predicts the status of chronic kidney disease based on various factor. Extensive data visualization and data preprocessing were done on the real-world data. The model were hyper-parameter tuned and cross validated to get the best result. Projects

PREDICT THE SENTIMENTS OF AMAZON CUSTOMER

This is a machine learning based project that predicts the sentiments of Amazon's customer using classification algorithm. NLP model and resampling methods were used on the real- world data set to build a better model.

PREDICT THE STOCK NEWS HEADLINES

This is an NLP model that predicts the future stock prices which was trained on real-world data. The model were hyperparameter tuned using grid search and cross validation were also performed.

PREDICT THE HOTEL BOOKING CANCELLATION

This machine learning project uses multiple algorithm to predict hotel booking cancellation based on real-world data. Detailed data analysis and data cleaning process is performed here.

SENTIMENT ANALYSIS WITH THREE PRE-TRAINED LANGUAGE MODEL (BERT, ELMO, ULM-FIT)

A person sentiments were analyzed in this language model where deep learning model BERT and ELMO, ULM-fit were applied on multiclass dataset.

Contact this candidate