Patrick Martins

As a recent graduate with a Bachelor's degree in Artificial Intelligence and Machine Learning, supplemented by a minor in MedTech and Healthcare, coupled with a solid IT background, I bring extensive knowledge and practical experience from innovative projects. Specifically, during my studies, I actively participated in relevant projects in the areas of Machine Learning, Data Science, and Python.

Stockmarket predictor

• Develop a model to predict daily market trends
• Prepare and analyze large datasets to ensure accurate predictions
• Apply Random Forests and RNNs for forecasting market direction
• Optimize models to enhance prediction accuracy

Fake News Detector

Bachelor thesis - Process development for AI use cases: 

• Develop a process to assist companies with AI integration
• Conduct an in-depth analysis of AI applications in a business context
• Structure and optimize complex processes for effective AI integration.

AI challenge - AMP-Parkinson's disease progression prediction: 

• Participate in a Kaggle challenge to predict Parkinson's disease progression using protein and peptide data.
• Experiment with models like LSTM, GRU, and RNN to enhance prediction accuracy.
• Utilize tools such as Weights & Biases for hyperparameter tuning
• Overcome data challenges through thorough data cleaning and detailed analysis.

NLP - Multiple choice question answering: 

• Train a model to answer multiple-choice questions using various approaches, including Word Embedding with Word2Vec, BERT, and ELECTRA
• Fine-tune the ELECTRA model with Weights & Biases to improve performance.

IIP - Recommender System: 

• Develop a recommendation system for bachelor’s theses as part of a team.
• Apply Word Embedding with Word2Vec to analyze user interests and knowledge, and generate personalized recommendations based on user profiles.