Roseline W.

Key Skills

Software

AWS SageMaker

Top Subject Matter

NLP complaint classification (customer feedback)

Predictive ML for tenant churn and late-payment risk

Legal Services & Contract Review

Top Data Types

Text

Document

Image

Top Task Types

Entity (NER) Classification

Classification

Freelancer Overview

Junior Data Scientist – ALX Africa (Project-Based). Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include N, A, and Python. Education includes AWS Cloud Practitioner, AWS re/Start (2024) and Certificate, ALX Africa (2024). AI-training focus includes data types such as Text and labeling workflows including Entity (NER) Classification and Classification.

Labeling Experience

Administrative Data Scientist – Elink Agencies Ltd

TextClassification

As an administrative data scientist, you built predictive machine learning models that categorize and identify high-risk individuals based on structured tenant/payment data. You performed exploratory data analysis to uncover drivers behind late payments and revenue leakage, then used modeling outputs to inform prioritization strategies. Your pipeline work supported automated ingestion, cleaning, validation, and downstream decision use in the organization. • Built predictive models in Python to forecast rental revenue, occupancy, and tenant churn. • Classified high-risk late-paying tenants to optimize follow-up prioritization. • Designed end-to-end data pipelines (Excel ➝ Python ➝ SQL) for ingestion and validation. • Performed statistical trend analysis on arrears, vacancies, and maintenance cost drivers.

2025 - Present

Junior Data Scientist – ALX Africa (Project-Based)

TextEntity (NER) Classification

In an internship role, you processed and modeled large volumes of text for NLP complaint classification tasks. You contributed to training and evaluation of models by preparing and using labeled/target categories for downstream predictions. Your work improved tagging accuracy significantly while supporting production-ready analysis and deployment preparation. • Processed 500K+ text records for NLP-based complaint classification. • Increased tagging accuracy by 60% through modeling and evaluation. • Built and assessed statistical models to investigate service inefficiencies and user behavior trends. • Delivered end-to-end ML workflows including data wrangling, modeling, evaluation, and presentation.

2024 - 2024

Education

Work History