This case study details Vivo Game Center's strategic redesign of its homepage, addressing content exposure, navigation efficiency, and visual immersion through a restructured layout, dynamic video integration, and a refined waterfall flow that together boost game activation rates and user engagement across devices.
From its inception as a simple game distribution platform to a sophisticated, multi‑layered recommendation architecture, this article details the background, early models, business growth, architectural evolution, caching strategies, GC optimization, rate limiting, experiment platform, multi‑path recall, dynamic tuning, and future intelligent enhancements of a game recommendation system.
This article chronicles the development, architectural challenges, and engineering solutions of a large‑scale game recommendation platform, covering background, initial models, business growth, caching strategies, GC optimization, rate‑limiting, fine‑grained operations, multi‑path recall, A/B testing, and future intelligent enhancements.
The article details the design, evolution, and operational practices of Vivo’s large‑scale game recommendation platform, covering its initial rule‑based model, layered strategy framework, multi‑level caching, GC tuning, rate‑limiting, fine‑grained A/B testing, multi‑path recall, dynamic exposure control, and future intelligent extensions.
This article explains why and how sample weighting is used in machine learning, illustrates YouTube's conversion of video watch time into sample weights to align with its commercial goals, and describes practical weighted‑logistic‑regression techniques applied to improve game recommendation systems.
An in‑depth overview of collaborative filtering, covering basic item‑based and user‑based methods, their limitations, advanced tricks such as IUF correction, popularity suppression, sliding‑window behavior sequencing, and a concrete implementation for game fashion recommendation, illustrating how to use CF as a baseline in early‑stage product development.
For deduplicating game recommendations in Redis, RoaringBitMap outperforms intset and Bloom filters by storing 300 auto‑incrementing game IDs in roughly 0.5 KB—over twice the compression of intset and far smaller than the 29 KB Bloom filter—thereby cutting memory use, latency, and hardware costs.
