Frank O.

Key Skills

Software

Don't disclose

Other

Top Subject Matter

Fake news detection and sentiment/NLP classification

SMS spam detection (supervised text classification)

AI-assisted content generation and chatbot-style support workflows

Top Data Types

Text

Image

Document

Top Task Types

Fine-tuning

Classification

Freelancer Overview

Fake News Detection Using Sentiment Analysis (MSc Dissertation). Brings 16+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Don't disclose and Other. Education includes Master of Science, University of Bradford (2024) and Bachelor of Science, Lagos State University (2010). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning, Classification, and Evaluation.

Labeling Experience

Lead Software Engineer (Volunteer) – MINARQ LTD (AI-assisted workflows)

OtherText

Built a system of AI-assisted content and planning workflows intended to reduce drafting time for website copy, branding text, and client-facing materials. Researched AI chatbot and automated support concepts to improve handling efficiency for common inquiries. Used AI tools to speed up idea generation and copy refinement, and explored automation opportunities for repetitive internal tasks that could reduce manual effort. • AI-assisted content drafting and refinement workflows for website/branding materials • AI chatbot and automated support research for common customer inquiries • AI-driven workflow automation exploration for internal repetitive tasks • Planning for AI-enabled digital product features with personalized assistance

2023 - Present

SMS Spam Classification (MSc Dissertation)

OtherTextClassification

Developed an SMS spam classification system in Python/Jupyter using the UCI SMS Spam Collection dataset. Performed exploratory data analysis, feature engineering, text preprocessing, and supervised model benchmarking to identify the best-performing classifier. Selected Extra Trees as the top model based on evaluation results, including reported accuracy and precision. • Text preprocessing and exploratory data analysis on SMS message dataset • Feature engineering and supervised classifier benchmarking • Model selection driven by evaluation metrics (accuracy/precision) • Training pipeline for text classification experiments

2023 - 2024

Fake News Detection Using Sentiment Analysis (MSc Dissertation)

Don't discloseTextFine-tuning

Built and fine-tuned a fake news detection system using BERT, T5, and RoBERTa across binary and multi-class classification tasks on large text datasets with custom preprocessing and class-imbalance handling. Evaluated multiple models and techniques to select approaches that performed best across complexity levels, including reporting accuracy and precision results. Applied NLP preprocessing steps such as cleaning and tokenisation and used feature reduction and benchmarking to drive model improvements. • Dataset preparation with custom preprocessing and handling of class imbalance • Model training/fine-tuning and benchmarking across BERT, T5, and RoBERTa • Evaluation across 2-way, 3-way, and 6-way classification, reporting accuracy/precision • NLP feature engineering and preprocessing (cleaning, tokenisation, feature reduction)

2023 - 2024

Education

Work History