Source Themes

Propose a cross-view approximation on Grassman manifold (CAGM) model to address inconsistencies within multiview adjacency matrices, feature matrices, and cross-view combinations from the two sources.

Propose an explainable stock earning framework via news factor analyzing model. Formalizing news into semantic graph and learn embedding of it as a more expressible factor. Stock earning foresting and explainable earning module are built via aggregating and utilizing news factor and numeric stock factor, achieving SOTA performance on A-stock dataset. A detailed report will be generated with the power of LLM.

Propose a simple but effective multigraph convolution networks, focusing on exacting edge-level and subgraph-level credible matrix to help cross-view interaction.

Propose an unsupervised framework to exact negative relationship between nodes, based on pseudo partial labels, and augment the original graph into a signed graph. Then signed graph neural networks are used for learning node embeddings, achieving state-of-the-art performance on both link prediction and node classification tasks.