Viridiana Sanchez

Logo

View My GitHub Profile

Data Analyst

Technical Skills: Python, SQL, R, Excel, Tableau, PowerBI, Azure

Education

Psychological Science, B.A. | California State University San Marcos

Certifications

Azure Data Scientist Associate | Microsoft Certification, November 2023

Work Experience

Data Analyst @ Division of Partnership, Prevention, and Services, Washington State DCYF

Case Resource Manager @ Department of Developmental Disabilities, Washington State DSHS

Data Analyst - Independent Contractor @ California State University San Marcos

Research Lab Manager @ California State University San Marcos

Culture + Intergroup Relations Lab

Publications

Sasha Y. Kimel, Jonas R. Kunst, Fatih Uenal, James Sidanius, Viridiana Sanchez Alcaraz. “They are what they eat”: Negative affect evoked by other’s food practices increase blatant outgroup dehumanization (Under Review).

Projects

Leverage Machine Learning to Create Recommendation Systems

Developed a content-based movie recommender using natural language processing that achieved a cosine similarity above 0.8, ensuring users receive highly relevant and personalized movie suggestions. Built a neural collaborative filtering model with a 0.8 RMSE, highlighting the system’s predictive performance in accurately recommending movies on predicted ratings derived from users’ historical rating behavior. View

rec1

Predictive Modeling with SciKit-Learn and TensorFlow

Implemented feature engineering strategies such as interaction terms and polynomial features to achieve a 20% reduction in Mean Squared Error across regression models to enhance precision. Leveraged the power of ensemble learning with random forest regression, achieving a 15% increase in R-squared values and significantly improving model explainability and predictive accuracy. View

ml

Decoding Data: A Step-by-Step Guide to Exploratory Data Analysis

Authored an EDA guide for dataset of 52,000 movies using Python, Numpy, Pandas, and Sci-Kit Learn that aims to increase users’ data literacy by 25% and improve their efficiency in exploring and interpreting data patterns through descriptive and inferential statistics, correlation matrix and visualizations. View