An Asymmetric Latent Factorization-of-Tensors Model for Relation Analysis

Latent Factorization-of-Tensors (LFT) models are an effective approach for relation analysis. Existing LFT models assume each mode of the target tensor corresponds to an entity set and the relationships between entity sets are bipartite graphs to explore the relationships among entities within a mode. However, when the topological structure of entities in a mode is known, for example, entities are ordered physical quantities, such as time or coordinates, the relations between such modes forms a more complicated structure, i.e., aligned bipartite networks, and existing LFT models cannot accurately capture this structure. This work is the first to recognize and analyze this issue, and proposes an Asymmetric Latent Factorization-of-Tensors (ALFT) model to address it. ALFT can model aligned bipartite networks in mode pairs of a tensor by imposing constraints between particular mode pairs in the tensor network. Experimental results on real-world datasets demonstrate the existence of this issue and confirm that the proposed ALFT model can effectively resolve it.

Many AI models analyze complex relational data by assuming that all entities are unordered groups. However, some entities, such as time or spatial coordinates, naturally contain internal structures. Ignoring these structures can reduce analysis accuracy. We developed a new tensor-based model called ALFT that can better capture these structured relationships. The model introduces special constraints to preserve the underlying topology between related modes in the data. Experiments on several real-world datasets show that ALFT consistently outperforms existing methods for relation analysis.