Predicting the TripleJ Hottest 100 With Python
summary
Python scripts scrape public Twitter and Instagram posts (wget, BeautifulSoup, platform APIs) to collect several thousand shared vote images and pages, feed them to tesseract OCR to extract text, clean it, then match strings with Levenshtein distance and aggregate results using multiprocessing worker pools. Replicates the Warmest 100 method and produces a ranked prediction soon after voting closes; accuracy improves with sample size but shows notable outliers where a high-ranked song isn’t present in the social sample.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 0:50Who is Justin
- 3:03The Toolchain
- 8:55The Data Processing
- 10:22The Data Analysis
- 13:26Multiprocessing
- 17:37Statistics
related talks
48mPythonic Polling Analysis and Comments on 2016's Polling Surprises
22mLet’s make the world a better place: Python for Social Good
52mPython as the multitool of data science
52mUsing Python to Get Out the Vote
31mA Recent History of Python Through Stack Overflow Questions and Answers
29mMissing the Apex: Predicting F1 Races with Python
20mPython In The Job Market
33mBuilding Data Visualisations in Python in Minutes