Gabriel Fair's Projects
Combined MCEdit & Pymclevel repository.
Cross-lingual metaphor detection.
Stuff I run on my Dell PowerEdge T430 Hypervisor/NAS
Attempt to recreate other's #TidyTuesday Submissions with an effort to learn from them and to improve them in some way
News and Culture Websites that I am archiving
The guts for computing data for OpenAlex. For more, see https://openalex.org/.
An open source artificial opinion analysis system
A data collection, enrichment and analysis system used in Orion search.
Open and cheap DIY IP-KVM based on Raspberry Pi
some silly pinterest-clone
Sample implementation of a politeness model, trained on the Stanford Politeness Corpus
The dashboard uses both structured and unstructured data, including Twitter data, news articles, and pricing information to give the user a general overview of current and historic Bitcoin activity. The historic price chart and current price, is useful for identifying where Bitcoin is currently sitting relative to the past. The chart showing a count of “#Bitcoin” and “#Ethereum” usage pulled from Twitter gives insight into the social media buzz surrounding this specific cryptocurrency, and some indication on the impact of any concurrent events. The word cloud generated from news sources can show general market sentiment from reputable news organizations which is useful if users of the dashboard have limited time to sift through various news websites. The Naïve Bayesian Classifier uses Google News to determine if the returns for the day will be positive or negative.
Tools to work with the big reddit JSON data dump.
Takes Reddit comments and Reddit submissions from files.pushshift.io and creates a nested collection out of it.
Simple Android app to add an icon to your share menu to archive a url on Archive.today and Archive.Is
Code for Scraper that was used for Shouting Into The Void, Gab.ai Dataset
Archives posts using archive.today
A monitor demo
Sparse and structured neural attention mechanisms
SparseMAP: differentiable sparse structure inference
Tutorial “Structural topic models for enriching quantitative text analysis” at the International Conference on Computational Social Science (IC2S2) in Amsterdam, NL, July 17th
Backtest 1000s of minute-by-minute trading algorithms with automated pricing data from: IEX, Tradier and FinViz. Datasets and trading performance automatically published to S3 for building AI training datasets for teaching DNNs how to trade. Runs on Kubernetes and docker-compose. >150 million trading history rows generated from +5000 algorithms. Heads up: Yahoo's Finance API was disabled on 2019-01-03 https://developer.yahoo.com/yql/
Summer 2017 Social Media Analytics Workshop Series
Tab Auto Close is a firefox plugin to autoclose tabs after certain amount of time.
Embed google news feed into your tableau dashboard
An auto downloader and uploader for TikTok videos.
Mine tweets with Python and parse/analyze with R and a Shiny Dashboard
VizNet is a repository providing real-world datasets that enable, among other things, (re)running empirical studies with higher ecological validity
A simple HTML5, YouTube and Vimeo player