Уровень 0 · материалов: 3
В кластер входят документы, описывающие количественные методы анализа и визуализации данных в сфере медиа, и не входят тексты, представляющие собой обсуждение потребительских предпочтений без опоры на массивы данных.
Общие признаки: использование больших данных, анализ трендов потребления контента, визуализация данных, статистический майнинг
Группа выше: Статистический анализ данных
Смысл: The author uses a massive XML dump of RuTracker's history to demonstrate a big data analytics workflow. By analyzing 15 years of data, the author concludes that while certain categories like movies and mobile software are declining, the platform remains active and continues to evolve.
Using R, Clickhouse, and Dataiku to analyze 15 years of RuTracker data, the author finds that while movie piracy is declining, the platform remains resilient and continues to grow in other categories.
Смысл: The text describes a data visualization project by Google Research called 'Music Timeline,' which uses Google Play Music data to show the evolution of music genre popularity over 64 years.
Google Research released 'Music Timeline,' an interactive infographic showing the popularity trends of music genres over the past 64 years using Google Play Music data.
Смысл: The main idea is that modern pop music relies on a limited and predictable set of harmonic patterns and chord progressions, which can be quantified through statistical data mining.
An analysis of 1,300 popular songs reveals that pop music follows highly predictable chord patterns, a fact mirrored in tools like Apple's GarageBand.