Prediction of Usage Probabilities of Shopping-Mall Corridors Using Heterogeneous Graph Neural Networks
Malik M Barakathullah, Immanuel Koh
TL;DR
The paper tackles predicting time-averaged usage probabilities on shopping-mall corridor edges by modeling each mall as a heterogeneous graph of shop and non-shop nodes connected by corridor paths. It introduces a heterogeneous Graph Neural Network autoencoder with a dedicated edge-predictor that leverages node-, edge-, and graph-level features and is trained on synthetically generated data via a probabilistic attraction model. Key contributions include (i) a data-generation pipeline grounded in shortest-path routing and probabilistic shop attributes, (ii) a representation-learning strategy for feature-poor non-shop nodes, and (iii) a deep, skip-connected GNN architecture that addresses oversmoothing while predicting edge-level targets. The approach demonstrates strong alignment with synthetic targets and shows potential for generalization to multi-mall datasets, with practical implications for estimating visibility, rents, and congestion management in mall design and operation. The work highlights how edge-level predictions, grounded in graph topology and domain-specific features, can inform planning decisions in built environments.
Abstract
We present a method based on graph neural network (GNN) for prediction of probabilities of usage of shopping-mall corridors. The heterogeneous graph network of shops and corridor paths are obtained from floorplans of the malls by creating vector layers for corridors, shops and entrances. These are subsequently assimilated into nodes and edges of graphs. The prediction of the usage probability is based on the shop features, namely, the area and usage categories they fall into, and on the graph connecting these shops, corridor junctions and entrances by corridor paths. Though the presented method is applicable for training on datasets obtained from a field survey or from pedestrian-detecting sensors, the target data of the supervised deep-learning work flow in this work are obtained from a probability method. We also include a context-specific representation learning of latent features. The usage-probability prediction is made on each edge, which is a connection by a section of corridor path between the adjacent nodes representing the shops or corridor points. To create a feature for each edge, the hidden-layer feature vectors acquired in the message-passing GNN layers at the nodes of each edge are averaged and concatenated with the vector obtained by their multiplication. These edge-features are then passed to multilayer perceptrons (MLP) to make the final prediction of usage probability on each edge. The samples of synthetic learning dataset for each shopping mall are obtained by changing the shops' usage and area categories, and by subsequently feeding the graph into the probability model. When including different shopping malls in a single dataset, we also propose to consider graph-level features to inform the model with specific identifying features of each mall.