JD's DPG image compression technique cuts image file size by about half, lowers CDN bandwidth by 50%, and delivers visual quality comparable to WebP across all browsers and platforms, thereby speeding rendering and improving shopping experience.
The article outlines the role of the content compliance domain in e‑commerce, detailing user‑facing issues, business responsibilities, challenges in detection and mitigation, and technical solutions such as comparable‑price models, large‑scale price prediction, and merchant outreach, while also offering personal growth advice for compliance engineers.
This article reviews the evolution of JD’s recommendation advertising ranking mechanism, covering its economic auction origins, challenges of multi‑material valuation, user interest uncertainty, and multi‑item auction fairness, and describes AI‑driven solutions such as deep auction models and reinforcement‑learning‑based ListVCG.
Lin Yuansheng, JD’s first post‑doctoral researcher, leverages his system‑engineering background and big‑data analytics to build models that identify black‑market accounts, improve network risk perception, and demonstrate how academic research can be rapidly applied to real‑world e‑commerce security challenges.
This article examines the evolution of Trip.com’s ticket product structure, detailing its transition from low to high‑level schema, the technical challenges faced across user, merchant, platform and R&D perspectives, and the systematic solutions—including structure integration, sales‑attribute standardization, and description‑attribute layering—that improved efficiency, data quality, and internationalization.
This article describes how Ctrip's R&D team applied deep‑learning models, BERT‑based embeddings, knowledge distillation, and term‑weighting techniques to enhance e‑commerce search intent recognition and term importance estimation, achieving high accuracy while meeting sub‑10 ms latency requirements.
This article presents a comprehensive technical case study of a three‑layer product‑tagging platform that addresses the challenges of fine‑grained operations, ensures real‑time tag updates, guarantees data consistency, and eliminates read bottlenecks through traffic separation, event‑driven processing, deduplication MQ, and multi‑level caching.
This article introduces multi‑task learning, reviews industry category‑prediction methods, and details Zhaozhuan's practical application of MTL to improve e‑commerce search intent understanding through hierarchical category, brand, and model prediction using RoBERTa and contrastive learning.
This article reviews the necessity, benefits, and practical implementations of multi-task learning in e‑commerce search, detailing model selection, architecture extensions such as ESMM and ESM², and future directions for handling user behavior sequences and multi‑objective optimization.
This article examines the characteristics of ZhaiZhai's B2B auction scenario, defines core pricing metrics, presents a step‑by‑step methodology for determining optimal starting prices, reviews early practices and their shortcomings, and details the current modular machine‑learning model architecture that improves transaction rates and reduces price premiums for second‑hand smartphones.
The article explains how an e‑commerce risk‑control team leverages OpenAI's CLIP model to generate image and text embeddings and stores them in the Milvus cloud‑native vector database to enable fast, scalable similarity searches for compliance verification and risk detection.
The article presents a comprehensive pricing solution for an e‑commerce platform that combines Bayesian statistical modeling, MCMC sampling, and causal inference (including Dragonnet) to achieve controllable, fine‑grained SKU‑level price estimation and optimization.
This article presents a comprehensive design for Zhuanzhuan's home‑page recommendation pipeline, detailing the system architecture, challenges of traffic efficiency and diversity, and a two‑stage solution that applies Proximal Policy Optimization reinforcement learning in the re‑ranking module and Determinantal Point Process optimization in the coarse‑ranking and traffic‑pool stages, followed by offline simulation, online deployment, and evaluation metrics.
This article introduces the fundamentals of recommendation systems, explains Zhuanzhuan's main recommendation scenarios and pipeline, and details three graph embedding methods—DeepWalk, node2vec, and EGES—along with their practical implementations in recall and coarse‑ranking stages.
The article presents recent statistics showing PHP still dominates web development, powering 77% of the top 10 million sites, most major CMS platforms, and a large share of e‑commerce solutions, while highlighting its use in large‑scale services such as Slack and Wikipedia.
This article explains the role of an order system in traditional e‑commerce enterprises, outlines its main functional modules and architectural design, and discusses future development directions, covering system relationships, core processes, inventory and payment strategies, state‑machine management, and scalability considerations.
The article explains how to design a scalable, highly available backend system capable of handling millions of requests per second by defining key performance metrics, estimating traffic with the 2/8 rule, and applying architectural patterns such as load‑balanced clusters, vertical service splitting, distributed caching, and database master‑slave replication, illustrated with a Taobao case study.
This article explains JD.com's end‑to‑end order fulfillment process, the concepts of 211/411 delivery promises, the mechanisms behind order splitting by warehouse, merchant and payment dimensions, and how monetary discounts are proportionally allocated across SKUs.
The article reviews Alipay's 2017 AR‑driven marketing campaign, surveys the rapid adoption of AR/VR/MR by major e‑commerce platforms, and explains the technical foundations and distinctions of vision‑based, marker‑less, and location‑based augmented reality, virtual reality, and mixed reality systems.
This article explains the role of an order system in traditional e‑commerce, outlines its main functional modules, describes the end‑to‑end order flow—including creation, payment, production, confirmation and completion—covers inventory reduction strategies, order splitting, reverse processes, state‑machine design, and discusses future evolution toward a unified order center.