How Feature-Induced Manifold Disambiguation Improves Video Tagging in Multi-View Learning

The paper "Feature‑Induced Manifold Disambiguation for Multi‑view Partial Multi‑label Learning" accepted at KDD 2020 introduces the MVPML framework and the FIMAN method, which leverage heterogeneous multimodal features to correct and supplement video tags, thereby boosting distribution efficiency in Alibaba Entertainment’s platforms.

Alibaba EntertainmentKDD 2020manifold disambiguation

0 likes · 3 min read

How Feature-Induced Manifold Disambiguation Improves Video Tagging in Multi-View Learning

Youku Technology

Feb 3, 2020 · Artificial Intelligence

Alibaba Entertainment Video Search Algorithms: Practice and Insights

In this talk, senior algorithm expert Ruo Ren outlines Alibaba Entertainment’s video‑search framework—covering basic relevance, ranking, and multimodal techniques that blend information retrieval, NLP, machine learning, and computer vision—using Youku as a case study to illustrate business needs, algorithmic challenges, and practical implementation solutions.

AI AlgorithmsAlibaba EntertainmentYouku

0 likes · 2 min read

Alibaba Entertainment Video Search Algorithms: Practice and Insights

Youku Technology

Jan 16, 2020 · Operations

How Youku Ensures Client Quality: Challenges, Build‑Based Testing, and Performance Strategies

In this technical talk, Alibaba Entertainment’s Youku client quality expert explains the major challenges of the client QA system, introduces a testing approach built around the packaging process, shares practical performance testing methods, and outlines the 2020 exploration and future plans.

Alibaba EntertainmentPerformance TestingTesting Automation

0 likes · 2 min read

How Youku Ensures Client Quality: Challenges, Build‑Based Testing, and Performance Strategies