levon003 Goto Github PK

followers: 1.0 following: 2.0 repos: 29.0 gists: 4.0

Name: Zachary Levonian

Type: User

Company: Digital Harbor Foundation

Bio: ML Engineer and Data Scientist. PhD from @grouplens. Previously: Amazon, CaringBridge, ConcertAI, MITRE

Twitter: zwlevonian

Location: Tacoma, WA, USA

Blog: https://levon003.github.io/

Zachary Levonian

Hi there! I'm a data scientist and researcher.

You can learn more about my research interests on my homepage.

My most recent release is llm-math-education, a Python package implementing retrieval-augmented generation for ChatGPT to support middle-school math tutoring.

Here are a few of my research prototypes:

HealthBlogRec: a prototype deep learning recommender system for health blogs.
ALSim: a simulation library that implements several state-of-the-art active learning algorithms.
wiki-ores-feedback: a web app and auditing analysis for vandalism detection classifiers on Wikipedia.

And here are some standalone analyses:

covid-data-analysis: Exploratory modeling with COVID-19 patient data. Explainable logistic regression models with Pandas, statsmodels, scikit-learn, and pytest.
biomarker-case-study: High-performance binary classification with patient demographic and biomarker data. Gradient Boosting Machines with Pandas and scikit-learn.

Zachary Levonian's Projects

alsim

ALSim is a Python library for iterative batch-mode active learning simulations.

biomarker-case-study

Modeling with patient biomarker data.

carleton-guest-lecture-june2019

Slides for a guest lecture at Carleton College.

covid-data-analysis

Basic modeling with COVID data; a good demonstration of my approach to data science.

cscw-caringbridge-interaction-network

Repository containing analysis code for research published in CSCW 2020.

dash-sample-apps

Apps hosted in the Dash Gallery

dolma-count-streamlit

Word counts for AI2 Dolma

dsinterviewqns

The Data Science Interview Book

grandtotem

Software implementation for the Grandtotem, a medium-fidelity communication prototype.

healthblogrec

A recommendation system for peer health blogs.

icwsm-cancer-journeys

Repository containing analysis code for research published in ICWSM 2020.

ipod-dataset-exploration

Quick exploration of the IPOD dataset.

kaggle-ga-customer-revenue

kaggle-predict-future-sales

Code for the Predict Future Sales competition on Kaggle

kaggle-titantic

levon003

Personal work and summary.

levon003.github.io

Zachary Levonian's personal website

ml-visualized

Miscellaneous code snippets and blog posts related to machine learning and statistics.

multithreaded-network-chatter

A networked Chat client/server implemented in C.

nlp-for-hci-workshop

Materials for a 2019 workshop on NLP for social computing/HCI researchers.

operations-mediawiki-config

⚙️ Configuration for Wikimedia Foundation wikis. This is a mirror from https://gerrit.wikimedia.org. See https://www.mediawiki.org/wiki/Developer_access for contributing.