Hey there!
I'm Ryan Sandan
Data Analyst III @ Juniper Networks
Data Science '22 @ UC Berkeley
Currently based in San Diego, CA
Brooklyn Nets #1 Fan
Get in touch LinkedIn
View my resume Resume
Tableau Profile Tableau Visualizations
Explored Yahoo's Fantasy API to analyze matchup data, free agency transactions, and player trends. Gained hands-on experience with OAuth 2.0 authentication, secure API token management, and web app deployment. Strengthened understanding of web hosting, API structures, and authentication workflows.
Future Enhancement: Scale the platform to support all Yahoo! NBA Fantasy leagues, expanding accessibility and usability for a broader audience.
Implemented clustering model to group NBA players based on traditional and advanced box score statistics to:
- Identify potential role player replacements
- Scout emerging talent
- Provide contingency planning for injuries and trades
Future improvement: Add NCAA player data.
With the 2nd worst loss in Nets Franchise history against the Celtics this past season (2023-24), I wanted to assess whether Nets GM Sean Mark, who is approaching 10 years of being the Nets GM, knows what he is doing and whether or not we're headed in the right direction.
Final Project for UC Berkeley's IEOR 142: Introduction to Machine Learning and Data Analytics
Team members: Ryan Sandan, Matthew Yep, Faizaan Merchant, Malik Ngom, Maryam Pourshariati
Tested Neural Network, Random Forest, and Logistic Regression model to predict NBA Allstars. Successfully predicted 88% (22/25) of 2023 Allstars after only a quarter of the season using Logistic Regression Model
Volunteered to develop HIPAA-compliant interactive map tracking high-risk community needs across the Triad region in North Carolina for Atrium Health, logged by Community Health Workers to optimize resource allocation.
Provided a personalized and engaging music discovery experience for Spotify users by recommending relevant songs aligned with their tastes. Implemented K-Means clustering, Elbow method, Silhouette score, and Euclidean distance function to optimally group songs into k = 5 distinct clusters based on Spotify API audio features
Performed exploratory data analysis on tweets from several high-public figures scraped from Twitter API. Findings include: top 10 most frequent hashtags and user mentions, distribution of tweets by time of day, and sentiment polarity scores.
As a course assistant for DATA88 I provided demos to introduce the data to students and provide exploratory data analysis. This is a demonstration of our final project analyzing covid19 data. Presentation link included at top of code.
My first personal website that I worked on during community college.