TY - JOUR
T1 - TSNN
T2 - A Non-Parametric and Interpretable Framework for Traffic Time Series Forecasting
AU - Liu, Bowen
AU - Lai, Haijian
AU - Lam, Chan Tong
AU - Dong, Junhao
AU - Ng, Benjamin
AU - Ke, Wei
AU - Im, Sio Kei
N1 - Publisher Copyright:
© 2026 IEEE.
PY - 2026/8/1
Y1 - 2026/8/1
N2 - Although many complex models were proposed to analyze time series data, some studies have demonstrated remarkable performance with simpler structures. A recent study proposed a non-parametric framework for 3D point cloud classification, which has the potential to be adapted for time series forecasting and enable interpretability. Inspired by the previous works, we present TSNN, a non-parametric and interpretable framework for traffic time series forecasting. TSNN consists of multiple layers that decouple the time series by matching the entries in a memory bank, where the memory bank is constructed using a similar matching process within the training set. It leverages the periodicity in traffic data to enhance forecasting accuracy while maintaining a simple model architecture. The proposed model operates without trainable parameters, preserving its inherent interpretability. In the experiments, TSNN achieves competitive performance compared to the typical deep learning models in four real-world traffic flow datasets. We also visualize the decoupling process to show the effectiveness of the components. Finally, we demonstrate the interpretability of the model and illustrate the contribution of each time step within the memory bank.
AB - Although many complex models were proposed to analyze time series data, some studies have demonstrated remarkable performance with simpler structures. A recent study proposed a non-parametric framework for 3D point cloud classification, which has the potential to be adapted for time series forecasting and enable interpretability. Inspired by the previous works, we present TSNN, a non-parametric and interpretable framework for traffic time series forecasting. TSNN consists of multiple layers that decouple the time series by matching the entries in a memory bank, where the memory bank is constructed using a similar matching process within the training set. It leverages the periodicity in traffic data to enhance forecasting accuracy while maintaining a simple model architecture. The proposed model operates without trainable parameters, preserving its inherent interpretability. In the experiments, TSNN achieves competitive performance compared to the typical deep learning models in four real-world traffic flow datasets. We also visualize the decoupling process to show the effectiveness of the components. Finally, we demonstrate the interpretability of the model and illustrate the contribution of each time step within the memory bank.
KW - Time series forecasting
KW - interpretability
KW - non-parametric model
KW - traffic flow forecasting
UR - https://www.scopus.com/pages/publications/105039242064
U2 - 10.1109/TKDE.2026.3692602
DO - 10.1109/TKDE.2026.3692602
M3 - Article
AN - SCOPUS:105039242064
SN - 1041-4347
VL - 38
SP - 5466
EP - 5479
JO - IEEE Transactions on Knowledge and Data Engineering
JF - IEEE Transactions on Knowledge and Data Engineering
IS - 8
ER -