This article explains the fundamentals of recommendation algorithms, focusing on user‑based collaborative filtering, similarity metrics, neighbor selection, scoring methods, practical implementation with the MovieLens dataset, and common challenges such as popularity bias and dirty data.
This article explains the unique characteristics of mother‑baby e‑commerce, describes a comprehensive big‑data platform architecture—including data collection, offline and real‑time computing, and recommendation algorithms—and shows how user profiling and personalized ranking dramatically improve conversion and user experience.
The Meizu Tech Open Day showcased the company's rapid evolution to a data‑driven mobile internet firm, detailing its DW1.0 and DW2.0 data‑warehouse architectures, recommendation pipelines, Spark adoption, and ELK‑based log analytics, while sharing practical lessons and future challenges.
This article explains the content‑based recommendation technique behind WeChat Moments ads, illustrates how user behavior is matched to ad attributes, and offers practical tips for influencing the system to display high‑value ads such as BMW.
This article walks through designing and implementing a full‑stack recommendation system for 58转转, covering product frameworks, user and item profiling, RFM modeling, personalized tagging, classification‑based and collaborative‑filtering approaches, and practical deployment tips.
This article explores how Meilishuo, China’s leading fast‑fashion discovery platform, tackles fragmented mobile attention by using AI‑powered personalization techniques—including user modeling, real‑time feedback, and tailored push notifications—to deliver highly relevant fashion recommendations and boost user engagement.
The article explains how JD.com leverages big‑data and personalized recommendation algorithms across PC and mobile platforms, detailing its recall and ranking models, efficiency analysis, weekly algorithm iterations, and future AI‑driven optimizations that together contribute about 10% of its orders.
This article explores Baidu Mobile's approach to personalization, detailing its recommendation algorithms, cloud‑side architecture, and front‑end engineering practices that together enable a highly tailored user experience across billions of users.
This article explains Meituan's end‑to‑end recommendation system architecture, data processing pipeline, candidate generation strategies, model training and online ranking techniques, illustrating how data, algorithms, and real‑time signals are combined to improve relevance and conversion.
This article examines how Baidu Mobile leverages personalized recommendation algorithms, rich user profiling, and a flexible cloud‑native architecture to deliver tailored search results and services, while also detailing the front‑end engineering practices that support its super‑app ecosystem.
This article traces the evolution of Weibo's recommendation architecture across three major phases—independent 1.0, layered 2.0, and platform‑centric 3.0—detailing the driving business and technical factors, architectural components, advantages, shortcomings, and key outcomes of each stage.
This article walks through the end‑to‑end process of building an offline recommendation system using Mahout, covering data collection, filtering, storage, various collaborative‑filtering algorithms, similarity measures, evaluation metrics, parameter tuning, AB testing, and spam‑fighting strategies.
This article traces the evolution of Weibo’s recommendation architecture across three major phases—independent 1.0, layered 2.0, and platform‑centric 3.0—detailing the environmental drivers, technical components, advantages, shortcomings, and key outcomes of each stage.
This article details the design, implementation, and performance of Lushan, a high‑throughput offline static data storage server built with libevent that supports dynamic library mounting, key‑value indexing, and efficient query handling for large‑scale recommendation workloads.
This article details Meituan's end‑to‑end recommendation system, covering its four‑layer architecture, data sources, candidate‑generation strategies, fusion methods, and both linear and non‑linear re‑ranking models, while highlighting practical optimizations like AB testing and online learning.
This article outlines Meituan’s end‑to‑end recommendation system, describing its data layer, candidate‑generation triggers, fusion strategies, and machine‑learning‑based ranking models—including collaborative filtering, location‑based, query‑based, graph‑based methods, and both linear and non‑linear models—while highlighting practical optimizations such as AB testing, real‑time behavior handling, and fallback strategies.
The article explains how Facebook built a large‑scale distributed recommendation system using Apache Giraph, collaborative filtering with matrix factorization, SGD and ALS algorithms, a novel work‑to‑work communication scheme, and performance optimizations that achieve ten‑fold speedups on billions of ratings.
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.
