Sara Abdelazim
Selected projects in data science, machine learning and NLP
Deep NLP for text deepfak detection
Text deepfak detection is the automated task of determining whether a piece of text contains artificially generated content. In this project, I utilized PyTorch to fine-tune BERT and Ro-BERT models for English text, AraBERT for Arabic text, and Multilingual BERT for multilingual capabilities.
Predicting Patient No-Shows: Enhancing Appointment Scheduling Efficiency
This project utilizes Support Vector Machine (SVM) and Decision Tree models to predict patient no-shows in a large medical center. The objective is to optimize appointment scheduling, resource allocation, and enhance efficiency in patient care delivery.
Exploring Drug Reviews through NLP: Sentiment Analysis and Classification
In this project I employed NLP techniques, including sentiment analysis, to analyze drug reviews and machine learning algorithms for drug category classification. EDA revealed trends like top conditions and drugs. Text preprocessing involved cleaning, tokenization, stop word removal, named entity recognition, and lemmatization.
-white){kind=icon}
Predicting DDoS Attacks: Uncovering Patterns, Enhancing Models
This project delved into predicting Distributed Denial of Service (DDoS) attacks through comprehensive Exploratory Data Analysis (EDA) to discern data patterns and trends. Two models were developed and evaluated: a baseline Linear Discriminant Analysis (LDA) model with hyperparameter tuning using GridSearch, and Random Forest Classifier aimed at robust attack classification.
-red){kind=icon}