Name: Oliver Zagorin
Type: User
Company: Catchafire
Bio: Business, Survey, and Data Analyst with experience gathering, analyzing and presenting data, to uncover hidden insights and represent them clearly.
Location: Washington, D.C.
Blog: https://www.oliverzagorin.com/
Oliver Zagorin's Projects
We used a dataset that included birth and personal data as well as Autism Spectrum Quotient test scores to train machine learning algorithms to predict autism. We used Logistic Regression, Neural Network Models and Keras Tuner with Random Oversampling to train one with 90% accuracy.
This project shows the creation of an SQL database from six related csv datasets, focused on an organization containing over 300,000 employees. I also conducted queries in order to obtain specific pre-defined results.
Trained and evaluated two supervised machine learning models using original and resampled data to identify 'healthy loan' and 'high risk loan' applicants from financial disclosures.
This was a team project focused on analyzing the relationship between the legal status of cannabis and crime rates in 2020. This includes the analysis, data visualizations, and a powerpoint presentation that discusses our insights and findings.
Analysis of crowdfunding campaigns to uncover actionable insights.
My team built an ETL pipeline using Python, Pandas, and regular expressions to extract and transform the data. After we transformed the data, we created four CSV files and used the CSV file data to create an Entity Relationship Diagram. Finally, we uploaded the CSV file data into a Postgres database.
Used K Means and PCA to analyze 42 cryptocurrencies in order to determine the effect of price changes over different periods of time.
This repository focuses on loading and performing exploratory analysis on a dataset in NoSQL and MongoDB which contains rating data for restaurants
Web Scraping HTML/CSS using BeautifulSoup on Mars News and Weather websites.
The analysis reveals that a small handful of microbial species (also called operational taxonomic units, or OTUs, in the study) were present in more than 70% of people, while the rest were relatively rare.
Analysis of over 34,000 businesses that received funding, to generate 184 Neural Network algorithm to predict effective allocation of funding.
We are performing two analyses using Python. The first analysis consists analyzing polling data for a small town, while the second consists of analyzing financial data.
Development and comparison of 12 machine learning models to predict autism as well as a discussion of the process.
Using Pandas and Jupyter Notebook, I created a report that analyzes Student and School data together in order to gain insights into how student education outcomes relate to a variety of school-level factors.
Outcome and insights analysis on "Multiple_year_stock_data" dataset, which contains stock data over the course of 3 years: 2018, 2019, and 2020. This dataset includes over 2.26 million individual entries.
Analysis of two datasets which relate to weather station information over the course of one year in Hawaii in 2016-2017. Analysis using SQLAlchemy ORM queries, and Pandas and Matplotlib in Python. A Flask API is created to store all the information.