Sherif Ahmed

A highly skilled professional with expertise in Python programming, database management, and data visualisation. Demonstrates strong capabilities in supervised and unsupervised learning, neural networks, and natural language processing. Adaptable to new technologies, with a focus on leveraging advanced data techniques to drive innovation and efficiency.

Offering a strong foundation in analytical thinking and problem-solving. Knowledgeable about statistical analysis, machine learning, and data visualisation. Strong communicator with collaborative approach and keen attention to detail. Ready to use and develop skills in Python, SQL, and data modelling in a Data scientist role.

Aspect-Based Sentiement Analysis on AWARE dataset

• Preprocessed noisy user-generated text using techniques like tokenization, lemmatization, and stopword removal.
• Built a custom aspect-opinion pair extraction pipeline using rule-based and ML-based methods.
• Fine-tuned transformer-based models (e.g., BERT, RoBERTa) for sentiment classification at the aspect level.

Disease Classification using Dermatology Dataset

• Developed multiple machine learning models to classify and cluster skin diseases using a real-world dermatology dataset.
• Applied classification techniques, including Random Forest and k-Nearest Neighbors, to identify disease types from medical features.
• Used clustering algorithms to discover natural groupings in the data and assess patterns in clinical attributes.
• Conducted model evaluation and comparison to understand performance trade-offs

Product Review Sentiment Analysis on Amazon Dataset

• Analyzed consumer product reviews using Python to extract insights on sentiment, themes, and customer behavior patterns.
• Cleaned and processed raw JSON data, handling missing values and irrelevant fields to prepare for analysis.
• Performed exploratory data analysis (EDA) on multiple products to uncover trends in ratings, review timing, and word usage.
• Applied Natural Language Processing (NLP) techniques such as tokenization, stop-word removal, stemming, and lemmatization to analyze review text.