Aya Amansour

Key Skills

Software

Appen

Clickworker

Data Annotation Tech

Top Subject Matter

Specializing in tagging and labeling Arabic text for machine learning applications.

Expertise in evaluating AI-generated content and ensuring cultural relevance in Arabic.

Proficient in annotating and transcribing Arabic audio for speech recognition models.

Top Data Types

Audio

Image

Text

Top Task Types

Bounding Box

Data Collection

Freelancer Overview

With a focus on both Arabic and English language challenges, I have extensive experience in data labelling and AI training. I contribute to the creation of precise and culturally appropriate AI models through my work in text annotation, transcription, and content evaluation. I am proficient in labelling for machine learning projects, content moderation, and speech-to-text transcription. I stand out for my meticulous attention to detail and multilingual work management skills, which guarantee dependable, high-quality data for AI training in a variety of industries.

Labeling Experience

Product Data Annotation Specialist

ClickworkerTextBounding BoxSegmentation

I worked on annotating product data for an e-commerce platform, ensuring accurate labeling of product descriptions, categories, and attributes. The tasks included: Identifying product-specific entities (NER) such as brand names, material types, and sizes. Classifying product descriptions into predefined categories. Evaluating and validating outputs from language models to ensure compliance with project guidelines. The project involved labeling over 50,000 product entries and achieving a 98% accuracy rate through rigorous quality control measures.

I worked on annotating product data for an e-commerce platform, ensuring accurate labeling of product descriptions, categories, and attributes. The tasks included: Identifying product-specific entities (NER) such as brand names, material types, and sizes. Classifying product descriptions into predefined categories. Evaluating and validating outputs from language models to ensure compliance with project guidelines. The project involved labeling over 50,000 product entries and achieving a 98% accuracy rate through rigorous quality control measures.

2023 - 2023

Education

Work History