ITPUB
Jul 23, 2020 · Artificial Intelligence
How Likee Scales Short‑Video Recommendations with Flink, Auto‑Stats, and Cache Tensor
This article details Likee's short‑video recommendation pipeline, covering the evolution of its feature‑engineering framework, the use of Flink for minute‑level statistical and second‑level session features, the integration of automatic statistical features into DNN models, multimodal feature extraction, and the cache‑tensor technique that dramatically improves online inference performance.
AIDeep LearningFlink
0 likes · 18 min read
