This article explains why Android apps suffer noticeable delays on cold start, breaks down the launch lifecycle, and provides practical optimizations—including reducing work in lifecycle callbacks, avoiding unnecessary delays, customizing window backgrounds, and using profiling tools—to dramatically cut startup time.
In this talk, Hulu’s research director Zhou Hanning explains the key factors influencing recommendation system performance, describes optimization goals, explores collaborative filtering, matrix factorization, and neural‑network approaches—including metadata‑driven transfer learning and cold‑start solutions for live streaming—and shares practical AI implementations that improve user experience and engagement.
After Facebook open‑sourced Redex, we explored its many optimizations and pitfalls, focusing on Interdex to reorder classes in the main dex, uncovering how class pre‑verification and hot‑patch instrumentation affect cold‑start performance, and sharing practical insights and remaining challenges for Android apps.
The QQ Music recommendation system tackles cold‑start by first mixing Chinese and English tracks, then builds a six‑dimensional user profile (content, social, scenario, crowd, time, blacklist) and tags songs with six attributes, using content‑based, collaborative, matrix‑factorization and neural‑network models plus implicit co‑listening links, while acknowledging that final wisdom still comes from human listeners.
This article explains the architecture of both system-wide and personalized recommendation modules, compares three recommendation strategies, details the use of Apache Mahout for collaborative filtering with Java code examples, and discusses cold‑start solutions within a Spark‑Hadoop stack.
This article outlines ten practical key points—including leveraging explicit and implicit feedback, hybridizing algorithms, handling temporal and geographic factors, exploiting social ties, solving cold‑start issues, optimizing presentation, defining clear metrics, ensuring real‑time updates, and scaling big‑data processing—to help engineers design effective intelligent recommendation systems.
This article surveys the fundamental misconceptions about personalized recommendation, distinguishes it from market segmentation and collaborative filtering, and then systematically presents ten critical research challenges—including data sparsity, cold‑start, scalability, diversity‑accuracy trade‑offs, system robustness, user behavior modeling, evaluation metrics, UI/UX, cross‑dimensional data integration, and social recommendation—each illustrated with examples and recent literature.
This article details Meituan's recommendation ecosystem, covering its key products, system goals, architecture, data pipelines, algorithms, cold‑start strategies, and the extensive ranking work—including modeling, sampling, bias removal, feature engineering, interleaving, and online learning—to dramatically boost user conversion.
