{
"name": "sisekelo sinyolo",
"occupation": "junior data scientist",
"interests": ["storytelling with data", "leveraging data for social impact"]
}
My Work
NLP Pipeline for isiNdebele - a low-resource language
Orange | Recommender system
Portfolio
development of Natural Language Processing (NLP) techniques tailored specifically for the isiNdebele language, enhancing digital accessibility and understanding of this underrepresented language.
Utilized a range of NLP methodologies including text scraping, data cleaning, language modeling, and semantic analysis to transform raw isiNdebele text data into actionable insights.
Contributed to the linguistic technology community by developing open-source tools and publishing detailed Jupyter notebooks that demonstrate each step of the NLP analysis, promoting further research and development in isiNdebele language processing.
Developed a robust subscription plan recommender system using advanced machine learning techniques, achieving an accuracy of 80.9% on unseen data.
Conducted thorough exploratory data analysis (EDA) and feature engineering, uncovering key insights and optimizing model performance with PyCaret and XGBoost.
Wine Market Data Analysis: Analyzed wine market trends and patterns, extracting actionable insights to guide marketing and sales strategies.
Tools used: Tableau, SQL
Accenture | Tableau analysis
About Me
I'm a junior data scientist working primarily in Python. Prior to joining the data science field, I worked in HR, Sales and Marketing. Although I'm new to the field, data understanding, grappling, visualisation, and communication have been an integral part of my work through all previous experiences.
I am currently completing a 7-month Data Science bootcamp at BeCode Brussels. Through the program we have built machine learning models for different company projects with Immoweb, Orange, Accenture. My most significant learning, beyond these projects, was exploring the isiNdebele language NLP pipeline through a personal project.
Aside my technology interests, I am an avid reader. I particularly love reading about history, economics, culture and sometimes, self-development. I am finalising my own book titled "My Identity, My Dreams" which is an autobiographical exploration of identity in our constantly changing world.
Technical skills
Python
AWS
Sci-kit learn
Git
FastAPI
Pandas
SQL
PowerBI
PostgreSQL
Numpy
Matplotlib
Tableau
© 2024 Sisekelo SInyolo