Text Annotation and Quality Review for Survey & Research Data
Worked on text labeling and quality review tasks involving survey and research data collected from users and patients. The project involved categorizing free-text responses into predefined classes, tagging responses based on sentiment and intent, validating question–answer pairs, and summarizing long text entries into concise, structured formats suitable for downstream analysis and AI training. Quality measures included strict adherence to labeling guidelines, double-checking ambiguous entries, maintaining consistency across labels, and flagging low-quality or incomplete responses for exclusion. I regularly reviewed labeled samples for accuracy, resolved edge cases through guideline-based decisions, and ensured datasets met required quality thresholds before final submission.