ABSTRACT
Region embedding is a primary task for a wide variety of urban-related downstream applications. However, many existing embedding techniques neglected the fact that the regions in a city have been developed differently by many factors such as planning policies, economic, and population mitigation. Such a spatial imbalance problem may result in a quite different region embedding to distinguish differences between regions, even though the regions could be similar in terms of the certain application tasks. In this paper, we propose a SPatial EMInence (SPEMI) model that normalizes region embeddings to mitigate the effects from spatial imbalance. In particular, we present a context-aware spatial feature, called spatial eminence, that measures a region's importance to its spatial context. The experimental results of store placement recommendation using real-world urban data show that SPEMI improves the performance of citywide region embeddings by up to 27.92%.
- Alexey Dosovitskiy and Lucas Beyer etc. 2021. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. In ICLR'21. OpenReview.net.Google Scholar
- Ziquan Fang, Dongen Wu, and Lu Pan etc. 2022. When Transfer Learning Meets Cross-City Urban Flow Prediction: Spatio-Temporal Adaptation Matters. In IJCAI'22. 2030--2036.Google Scholar
- Bin Guo, Jing Li, and Vincent W Zheng etc. 2018. CityTransfer: Transferring Inter- and Intra-City Knowledge for Chain Store Site Recommendation Based on Multi-Source Urban Data. UbiComp'18 1 (1 2018), 1010--1021. Issue 4.Google ScholarDigital Library
- Kaiming He, Xiangyu Zhang, and Shaoqing Ren etc. 2016. Deep residual learning for image recognition. In CVPR'16. 770--778.Google ScholarCross Ref
- Porter Jenkins and Ahmad Farag etc. 2019. Unsupervised Representation Learning of Spatial Data via Multimodal Embedding. In CIKM'19. 1993--2002.Google ScholarDigital Library
- Jiabao Jin and Peng Cheng etc. 2022. GridTuner: Reinvestigate Grid Size Selection for Spatiotemporal Prediction Models. In ICDE'22. IEEE, 1193--1205.Google Scholar
- Dichao Li and Zhiguo Gong. 2020. A Deep Neural Network for Crossing-City POI Recommendations. TKDE'20 (2020), 3536--3548.Google Scholar
- Yiding Liu, Kaiqi Zhao, and Gao Cong. 2018. Efficient Similar Region Search with Deep Metric Learning. In KDD'18. 1850--1859.Google Scholar
- Anthony Ortiz, Caleb Robinson, and Dan Morris etc. 2020. Local context normalization: Revisiting local normalization. In CVPR'20. 11276--11285.Google ScholarCross Ref
- Chang Sheng, Yu Zheng, and Wynne Hsu etc. 2010. Answering Top-k Similar Region Queries. In DASFAA'20. 186--201.Google Scholar
- Ashish Vaswani, Noam Shazeer, and Niki Parmar etc. 2017. Attention is All you Need. In NeurIPS'17. 5998--6008.Google Scholar
- Senzhang Wang, Hao Miao, and Jiyue Li etc. 2022. Spatio-Temporal Knowledge Transfer for Urban Crowd Flow Prediction via Deep Attentive Adaptation Networks. IEEE Trans. Intell. Transp. Syst. 23, 5 (2022), 4695--4705.Google ScholarDigital Library
- Max Welling and Thomas N Kipf. 2016. Semi-supervised classification with graph convolutional networks. In ICLR'17. OpenReview.net.Google Scholar
- Mingyang Zhang, Tong Li, and Yong Li etc. 2020. Multi-View Joint Graph Representation Learning for Urban Region Embedding. In IJCAI'20. 4431--4437.Google Scholar
- Zuoquan Zhao, Roger R Stough, and Ning Li. 2003. Note on the Measurement of Spatial Imbalance. Geographical Analysis 35 (2003), 170--176. Issue 2.Google ScholarCross Ref
Index Terms
- SPEMI: normalizing spatial imbalance with spatial eminence transformer for citywide region embedding
Recommendations
STcorr: An IDL code for image based normalization of lapse rate and illumination effects on nighttime TIR imagery
Thermal infrared imagery (TIR) is a useful tool to detect and quantify the surface temperature anomalies associated with geothermal fields. Accurate detection of anomalies in surface temperature is an important aspect of geothermal research. Although ...
Biometric iris recognition using radial basis function neural network
AbstractThe consistent and efficient method for the identification of biometrics is the iris recognition in view of the fact that it has richness in texture information. A good number of features performed in the past are built on handcrafted features. ...
Large-Scale Evaluation of Multimodal Biometric Authentication Using State-of-the-Art Systems
We examine the performance of multimodal biometric authentication systems using state-of-the-art Commercial Off-the-Shelf (COTS) fingerprint and face biometric systems on a population approaching 1,000 individuals. The majority of prior studies of ...
Comments