mlsanityData to 14 Oct 2026

FTSCommDetector: Discovering Behavioral Communities through Temporal Synchronization

arXiv 2510.00014 ↗First posted 17 Sep 2025cs.SI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Why do trillion-dollar tech giants AAPL and MSFT diverge into different response patterns during market disruptions despite identical sector classifications? This paradox reveals a fundamental limitation: traditional community detection methods fail to capture synchronization-desynchronization patterns where entities move independently yet align during critical moments. To this end, we introduce FTSCommDetector, implementing our Temporal Coherence Architecture (TCA) to discover similar and dissimilar communities in continuous multivariate time series. Unlike existing methods that process each timestamp independently, causing unstable community assignments and missing evolving relationships, our approach maintains coherence through dual-scale encoding and static topology with dynamic attention. Furthermore, we establish information-theoretic foundations demonstrating how scale separation maximizes complementary information and introduce Normalized Temporal Profiles (NTP) for scale-invariant evaluation. As a result, FTSCommDetector achieves consistent improvements across four diverse financial markets (SP100, SP500, SP1000, Nikkei 225), with gains ranging from 3.5% to 11.1% over the strongest baselines. The method demonstrates remarkable robustness with only 2% performance variation across window sizes from 60 to 120 days, making dataset-specific tuning unnecessary, providing practical insights for portfolio construction and risk management.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 33 references that are in the map. First posted on arXiv 17 Sep 2025.

Builds on

20 of 33 references.

References · 20

  1. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 931
  2. 2020Temporal Graph Networks for Deep Learning on Dynamic Graphscited by 530
  3. 2020Masked Label Prediction: Unified Message Passing Model for Semi-Supervised Classificationcited by 506
  4. 2019Predicting Dynamic Embedding Trajectory in Temporal Interaction Networkscited by 448
  5. 2019EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphscited by 447
  6. 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 346
  7. 2023Towards Better Dynamic Graph Learning: New Architecture and Unified Librarycited by 157
  8. 2019Attributed Graph Clustering: A Deep Attentional Embedding Approachcited by 149
  9. 2020Structural Deep Clustering Networkcited by 145
  10. 2022ROLAND: Graph Learning Framework for Dynamic Graphscited by 132
  11. 2023Koopa: Learning Non-stationary Time Series Dynamics with Koopman Predictorscited by 113
  12. 2023Time Series as Images: Vision Transformer for Irregularly Sampled Time Seriescited by 79
8 more
  1. 2023Cluster-guided Contrastive Graph Clustering Networkcited by 60
  2. 2023Deep Temporal Graph Clusteringcited by 28
  3. 2023DGCLUSTER: A Neural Framework for Attributed Graph Clustering via Modularity Maximizationcited by 21
  4. 2024LSEnet: Lorentz Structural Entropy Neural Network for Deep Graph Clusteringcited by 21
  5. 2023Graph-based Time Series Clustering for End-to-End Hierarchical Forecastingcited by 17
  6. 2022VGAER: Graph Neural Network Reconstruction based Community Detectioncited by 14
  7. 2023Time Series Clustering With Random Convolutional Kernelscited by 6
  8. 2020Amortized Probabilistic Detection of Communities in Graphscited by 4

Built on by

Not cited yet.