Ahmed E.

Key Skills

Software

Don't disclose

Top Subject Matter

Mental health support and NLP emotion classification

OCR and structured text extraction from images

Supervised ML model training/evaluation workflows

Top Data Types

Text

Image

Top Task Types

Emotion Recognition

Text Generation

Classification

Freelancer Overview

Graduation Project: AI-Powered Mental Health Detection (emotion classification via Bi-LSTM). Core strengths include Don't disclose, PyTesseract, and scikit-learn. Education includes Bachelor of Science, Misr University for Science and Technology (2025) and Diploma, Route Academy (2025). AI-training focus includes data types such as Text, Image, and Computer Code and labeling workflows including Emotion Recognition, Text Generation, and Classification.

Labeling Experience

OCR Text Extractor project using PyTesseract

ImageText Generation

Implemented an OCR-based text extraction workflow to turn image inputs into structured text. The project uses PyTesseract to extract text from uploaded images and supports converting the result into a usable text output. This constitutes annotation-oriented processing where visual text is transformed into machine-readable labeled text. • OCR extraction from uploaded images using PyTesseract • Conversion of extracted content into structured text outputs • Handling of varied image inputs for more reliable extraction • Output validation aligned with downstream use of extracted text

2024 - 2024

Graduation Project: AI-Powered Mental Health Detection (emotion classification via Bi-LSTM)

Don't discloseTextEmotion Recognition

Built an AI/NLP system to analyze user text in real time and classify emotional states. The model outputs categories such as anxiety, depression, or normal based on a custom Bi-LSTM deep learning pipeline. Labeling was performed implicitly through preparing/structuring training targets for emotion classes and evaluating the classifier on labeled examples. • Emotion category classification (anxiety/depression/normal) from user-provided text • Preparation/structuring of labeled datasets for model training • Model evaluation using labeled test instances and error metrics • Integration of NLP features to improve classification accuracy

2024 - 2024

Real-time ML application projects (Streamlit apps for classification/regression/filtering)

Classification

Developed end-to-end machine learning applications that support classification, regression, and filtering using structured datasets. While the resume does not specify manual annotation, the work includes preparing labeled targets and applying supervised learning workflows. These projects involve generating model outputs against labeled training/evaluation data and presenting results with error metrics. • Building supervised ML pipelines for classification and regression • Preprocessing of labeled datasets (missing values, scaling, encoding) • Interactive analysis and visualization of model performance • Deployment of user-facing apps for running predictive workflows

2023 - 2023

Education