NetEase Media Technology Team
Author

NetEase Media Technology Team

NetEase Media Technology Team

55
Articles
0
Likes
121
Views
0
Comments
Recent Articles

Latest from NetEase Media Technology Team

55 recent articles
NetEase Media Technology Team
NetEase Media Technology Team
Apr 11, 2022 · Artificial Intelligence

Multimodal Video Tagging: Challenges and a Two‑Stage Recall‑Ranking Solution

To tackle the massive, multimodal tagging challenge of short‑video platforms—characterized by a huge long‑tail tag set, sparse annotations, and uneven modality contributions—the authors propose a two‑stage recall‑ranking system that first retrieves candidates via text, visual, audio and classification cues, then refines them with contrastive learning and extensive hard‑negative sampling, achieving 0.884 tag accuracy in a real‑world news video recommender.

EmbeddingMultimodal LearningRecommendation Systems
0 likes · 12 min read
Multimodal Video Tagging: Challenges and a Two‑Stage Recall‑Ranking Solution
NetEase Media Technology Team
NetEase Media Technology Team
Dec 2, 2021 · Cloud Native

Why NetEase Media Adopted Go and Built the NGO Cloud‑Native Web Framework

Facing high memory usage and slow builds with Java SpringBoot, NetEase Media switched to Go in 2020, rebuilt core services, and created the cloud‑native ngo framework to deliver lighter containers, faster compilation, built‑in business libraries, tracing and monitoring, now running in production while hiring engineers.

Cloud NativeGoMicroservices
0 likes · 8 min read
Why NetEase Media Adopted Go and Built the NGO Cloud‑Native Web Framework
NetEase Media Technology Team
NetEase Media Technology Team
Aug 25, 2021 · Cloud Native

NetEase Media Container Platform Construction: Cloud Native Implementation Experience and Best Practices

NetEase Media details its year‑long journey building a cloud‑native container platform—covering foundational concepts, a robust infrastructure framework, Kubernetes deployment, solutions to pre‑containerization challenges, and practical best practices such as graceful shutdowns, health probes, and resource‑limit configurations.

Cloud NativeContainer TechnologyDevOps
0 likes · 32 min read
NetEase Media Container Platform Construction: Cloud Native Implementation Experience and Best Practices
NetEase Media Technology Team
NetEase Media Technology Team
Jul 23, 2021 · Frontend Development

CSS Animation Techniques for Medal Unlock and Gyroscope Effects

The article explains how to create a seamless medal‑unlock animation using layered SVG clipPath tricks and a four‑step CSS process, and how to build an interactive gyroscope effect with deviceorientation events, while emphasizing performance‑friendly practices such as using transform, opacity, requestAnimationFrame, and avoiding reflow‑inducing properties.

3D effectsSVG clipPathWeb animation
0 likes · 10 min read
CSS Animation Techniques for Medal Unlock and Gyroscope Effects
NetEase Media Technology Team
NetEase Media Technology Team
Jul 21, 2021 · Mobile Development

Componentization Practices in NetEase News iOS Application

The NetEase News iOS team adopted a three‑layer componentization strategy—foundation, common, and business—standardizing code conventions, Git workflow, and CI‑driven reviews, encapsulating third‑party libraries, managing dependencies via injection and mediators, and distributing binary pods, which reduced the main project to a single file while expanding reusable components to over 250, achieving low coupling and high cohesion.

Component ArchitecturePoddependency management
0 likes · 19 min read
Componentization Practices in NetEase News iOS Application
NetEase Media Technology Team
NetEase Media Technology Team
Apr 13, 2021 · Artificial Intelligence

Applying BERT for News Timeliness Classification at NetEase

The article describes how NetEase adapts a pre‑trained BERT model to classify news articles into ultra‑short, short, or long timeliness categories by combining rule‑based strong and weak time cues, key‑sentence extraction, domain‑embedding fusion and multi‑layer semantic aggregation, achieving accurate and interpretable predictions for its platform.

BERTModel FusionNLP
0 likes · 12 min read
Applying BERT for News Timeliness Classification at NetEase