Community Blog Series Cloud Knowledge Discovery on KDD Papers
+ New Series

Cloud Knowledge Discovery on KDD Papers

This series features 12 papers from the annual KDD conference, which includes explanations from 12 Alibaba experts around Alibaba Cloud technologies in practice.

Privileged Features Distillation at Taobao Recommendations

10 Alibaba Cloud experts discuss feature input in prediction tasks and using PFD to maximize privileged features, particularly with Taobao recommendations.

M2GRL: A Multi-Task Multi-View Graph Representation Learning Framework for Web-Scale Recommender Systems

5 researchers propose the usage of a multi-task multi-view graph representation learning framework to learn node representations from multi-view graph...

KDD 2020: Semi-Supervised Collaborative Filtering by Text-Enhanced Domain Adaptation

Hear 3 experts from the Alibaba Group and 2 researchers from Tsinghua University talk about recommendation algorithms and machine learning.

GCC: Graph Contrastive Coding for Graph Neural Network Pre-training

This article introduces Graph Contrastive Coding (GCC), pre-training framework that uses the contrastive learning method to pre-train graph neural networks.

Comprehensive Information Integration Modeling Framework for Video Titling

This article talks about the Gavotte model, which is used to automatically generate titles for buyer-uploaded videos in e-commerce scenarios.

Learning Stable Graphs from Heterogeneous Confounded Environments

We present the SGL learning framework to learn stable graph structures from heterogeneous confounded environments.

Disentangled Self-supervision in Sequential Recommenders

This paper presents a new training method that allows the recommender to learn the user's next intention and more intentions to come.

Understanding Negative Sampling in Graph Representation Learning

6 researchers explain benefits and challenges of graph representation learning.

Controllable Multi-Interest Framework for Recommendation

6 researchers propose a controllable multi-interest framework for recommendation, which is used to recommend items based on a user's click sequence.

A Dual Heterogeneous Graph Attention Network to Improve Long-Tail Performance for Shop Search in E-Commerce

This blog talks about Taobao's shop search service, and how Taobao successfully improved long-tail search performance with DHGAN.

Prediction and Profiling the Audience Competition for Online TV Series

Five researchers explore the audience competition for online TV series by providing the competitiveness definition, algorithm design, and experimental comparison.

KDD 2020: Learning to Generate Personalized Query Auto-Completions with a New Approach

6 researchers explain the typical problems of query auto-completion with search engines and why the multi-view/multi-task attentive approach can be the solution.

Loading more…