Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a novel method for constructing dynamic hypergraph representations from multivariate time series data without relying on prior knowledge. The approach uses community detection on the time series, transforms the communities into a hypergraph via a clique-based technique informed by attention weights, and then uses a Dynamic Hypergraph Attention Convolution Network (DHACN) for time series prediction. Experiments on various datasets demonstrate the effectiveness of the learned hypergraph representations for capturing high-order relationships in time series data.
Discover hidden relationships in your time series data: this method automatically builds dynamic hypergraphs to capture complex, high-order dependencies without needing any prior knowledge.
Hypergraphs have the capacity to capture higher-dimensional relationships among entities across various domains, making them a subject of growing interest within the research community for understanding the structure and dynamics of complex systems. However, a key challenge is the derivation of hypergraph representations from time series data in situations where the structure of the hypergraph is limited or absent. In this study, we propose a model that constructs a dynamic hypergraph representation for multivariate time series without relying on prior knowledge of the data. This is achieved by applying community detection to the time series and transforming the resulting communities, obtained through an attention mechanism, into a hypergraph using a clique-based technique. Hypergraph representations are derived from different time series datasets, and the resulting hypergraphs are then used by a Dynamic Hypergraph Attention Convolution Network (DHACN) for multivariate time series predictions. This research advances the field of hypergraph representation by introducing a novel approach that is better suited to uncover high-order relationships without prior knowledge.