Gabriel Martinez Email & Phone Number

Atlanta, Georgia, United States

Remote

- Learned how to scrape websites and clean subsequent data.- Developed crucial Python skills to build and tweak statistical models.- Learned how to build, train, and assess several machine learning models.- Learned and built boosted models.Projects:House Price Prediction- Built several models, including linear regression, elastic regression, and random forest regression models, to predict house prices based on features of the property.- Identified and selected the features that contributed the most to house pricing.- Cross-validated all models and used GridSearch to optimize hyperparameters.- Achieved an R-squared of 0.90 with the random forest regression model.Customer Segmentation- Used several clustering techniques, including agglomerative clustering and K-Means, to profile unique groups of customers in a company's dataset.- Cleaned the customer dataset and engineered several features to better profile customers.- Reduced the dimensionality of the data with PCA and t_SNE.- Found 4 unique clusters of customers that varied on several variables including income, age, expenses, items bought, and others.Amazon Review Sentiment Analysis- Trained a random forest classifier and convolutional neural network to perform sentiment analysis on a product's reviews of the user's choice.- The models were trained on thousands of labeled reviews.- Data preprocessing involved tokenization and lemmatization followed by eliminating punctuation, non-alpha characters, and stopwords.- The random forest classifier was trained on data processed with TF-IDF, which is a statistic that measures the importance of a given word in a corpus.- The convolutional neural network was trained on more training data because computation allowed for it.- The convolutional neural network was generally better across the board (~0.93 vs. ~0.83), including statistics such as accuracy, precision, recall, and ROC-AUC scores.

Atlanta, Georgia

- Leveraged large clinical datasets to analyze relationships between gray matter and relevant symptoms.- Cleaned data and conducted QA by developing scripts with MATLAB and R.- Created multivariate linear mixed-effects regression models using R.- Normalized abnormal data distributions with several transformation techniques.- Used several machine learning techniques, such as support vector regression, to predict symptoms and diagnoses in clinical datasets.

Atlanta, Georgia

- Assisted in the research, design, and administration of in-lab and remote experiments.- Conducted literature searches to create/support hypotheses.- Created and maintained a database of participants with Access and Excel.- Extracted, cleaned, and analyzed survey data from Qualtrics.

Master'S Degree, Neuroscience

Bachelor Of Science - Bs, Psychology