This website is a platform that showcases my skills and accomplishments in data science and data analysis.
This project explores predicting U.S. presidential election results at the county level using key data from the American Community Survey (ACS). Instead of relying on campaign events or social media, it focuses on core factors like income, age, gender, marital status, race, citizenship, language spoken at home, education, and employment. By analyzing these basic community traits, this project aims to see if election outcomes can be accurately forecasted.
Learn moreThe goal of this project is to demonstrate the use of prompt engineering with GPT. I Employed OpenAI's GPT-3.5 to analyze and score Amazon product reviews.
Learn moreThe goal of this project is to demonstrate the use of Hugging Face Transformers and BERT for sentiment analysis, and to compare the performance with Naive Bayes for predicting product ratings on Amazon based on customer reviews.
Learn moreThis project develops a fine-tuned BERT model to create embeddings of Amazon product reviews, with t-SNE visualization revealing structure and relationships within the data.
Learn moreThe goal of this project is to predict the rating of a product on Amazon based on the text of a customer's review. We have a dataset of customer reviews for different products on Amazon, each associated with a rating ranging from 1 to 5.
Learn moreThe study investigates the effects of maternal stress volatility during pregnancy on child health outcomes using modern biostatistics and data science techniques.
Learn moreIn this project, we analyze the data from the Cherry Blossom 10-mile running race from 1973-2022, which took place in Washington DC, to elucidate the connection between age and physical fitness.
Learn moreIn this food recommendation system project, I utilized feature engineering and scaling to preprocess nutritional data. By independently implementing K-means clustering and KNN regression algorithms from scratch.
Learn moreIn this innovative project, I collected diverse images of US coins using my smartphone and labeled them with bounding boxes through the LandingLens platform. I independently trained a deep neural network for object detection.
Learn moreIn this collaborative project, our team employed dimensionality reduction techniques, such as principal component analysis and recursive feature elimination, developing predictive models using various approaches.
Learn moreIn this USGS project, we investigated Klamath River's thermal stratification by examining weather and river flow data. We employed generalized additive models, random forest models, and K-means clustering to uncover patterns.
Learn moreIn this collaborative project, we examined the impact of the COVID19 pandemic on transgender women's health care experiences,with a focus on their financial health and gender affirming care using statistical analysis and methods such as Network Analysis and Graph Theory.
Learn more