Youtube视频搜索

1. requirements

• What is the primary (business) objective of the search system?

• What are the specific use cases and scenarios where it will be applied?

• What are the system requirements (such as response time, accuracy, scalability, and integration with existing systems or platforms)?

• What is the expected scale of the system in terms of data and user interactions?

• Is their any data available? What format?

• Can we use video metadata? Yes

• Do we need consider Personalization? not required

• How many languages needs to be supported?

2. pipeline

3. data collection

4. feature

5. model

text

video

loss

6. Evaluation

• Offline

  • Precision@k, mAP, Recall@k, MRR

  • we choose MRR (avg rank of first relevant element in results) due to the format of our eval data <video, text> pair

• Online

  • CTR: problem: doesn't track relevancy, click baits

  • video completion rate: partially watched videos might still found relevant by user

  • total watch time

  • we choose total watch time: good indicator of relevance

7. deployment and prediction service

• A/B test

• Scaling

8. monitoring and maintenance

reference

• Embedding-based Retrieval in Facebook Search

• Que2Search: Fast and Accurate Query and Document Understanding for Search at Facebook

• System Design for Recommendations and Search

• Applying Deep Learning To Airbnb Search

• https://cloud.google.com/vertex-ai/docs/vector-search/overview?hl=en

• 大众点评搜索相关性技术探索与实践

• 预训练技术在美团到店搜索广告中的应用

• 深度学习在搜索业务中的探索与实践

• 美团搜索粗排优化的探索与实践

• 美团点评旅游搜索召回策略的演进

• How image search works at Dropbox

• 京东 Towards Personalized and Semantic Retrieval: An End-to-End Solution for E-commerce Search via Embedding Learning

• 百度凤巢大模型与搜索广告满意度设计与实践

• POI 识别在飞猪搜索的探索与实践

• 大众点评内容搜索算法优化的探索与实践

• Under the hood: Photo Search

• 万字浅析视频搜索系统中的多模态能力建设 - 徐土豆的文章 - 知乎