Lyuwen Wu

With the enrichment of user-item interactions, Graph Neural Networks (GNNs) are widely used in recommender systems to alleviate information overload. Nevertheless, they still suffer from the cold-start issue. Knowledge Graphs (KGs), providing external information, have been extensively applied in GNN-based methods to mitigate this issue. However, current KG-aware recommendation methods suffer from the knowledge imbalance problem caused by incompleteness of existing KGs. This imbalance is reflected by the long-tail phenomenon of item attributes, i.e., unpopular items usually lack more attributes compared to popular items. To tackle this problem, we propose a novel framework called EditKG: Editing Knowledge Graph for Recommendation, to balance attribute distribution of items via editing KGs. EditKG consists of two key designs: Knowledge Generator and Knowledge Deleter. Knowledge Generator generates attributes for items by exploring their mutual information correlations and semantic correlations. Knowledge Deleter removes the task-irrelevant item attributes according to the parameterized task relevance score, while dropping the spurious item attributes through aligning the attribute scores. Extensive experiments on three benchmark datasets demonstrate that EditKG significantly outperforms state-of-the-art methods, and achieves 8.98% average improvement. The implementations are available at https://github.com/gutang-97/2024SIGIR-EditKG.

The large volumes of data on the Internet provides new opportunities for scientific discovery, especially promoting data-driven open science research. However, due to lack of accurate semantic markups, finding relevant data is still difficult. To address this problem, we develop a one-stop dataset service called DataExpo and propose a deep learning method for automatic metadata ingestion. In this demo paper, we describe the system architecture, and how DataExpo facilitates dataset discovery, search and recommendation. Up till now, DataExpo has indexed over 960,000 datasets from more than 27,000 repositories in the context of Deep-time Digital Earth Program. Demo visitors can explore our service via https://dataexpo.acemap.info.