基于半监督学习的小样本多维度城市指标预测方法、系统、终端及存储介质
By employing a semi-supervised learning-based method for predicting urban indicators in small samples and using a deep feature encoder and an iteratively trained model, this method addresses the issues of low accuracy and stability in urban indicator calculations in existing technologies, achieving high-precision prediction under small sample conditions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHENZHEN UNIV
- Filing Date
- 2026-04-24
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies rely heavily on large-scale labeled samples in urban indicator calculations, resulting in low accuracy and stability of urban indicator calculations under conditions of scarce labeled samples. Existing methods lack a systematic approach to deeply integrate semi-supervised learning mechanisms with the characteristics of urban spatial data.
A small-sample, multi-dimensional urban indicator prediction method based on semi-supervised learning is adopted. By acquiring multiple modal data and mapping them to spatial units, physical, functional, and structural features are extracted using a deep feature encoder. Labeled and unlabeled datasets are constructed, and deep neural networks and deep ensemble tree models are used for alternating iterative training. Multiple reset reliability verification constraints are constructed to filter pseudo-labels, generate a target labeled dataset, and finally construct a target prediction model.
It significantly reduces the reliance on labeled data, achieves high-precision urban indicator prediction under small sample conditions, and improves the accuracy and stability of urban indicator calculation.
Smart Images
Figure CN122087370B_ABST