Land Usage Prediction Using Graph Convolutional Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting regional land usage properties are inadequate in accurately forecasting changes over time, especially with the dynamic nature of urbanization and industrialization, which affects urban planning and other fields.
Innovation Solution
A method involving the acquisition of human interaction information between regions, updating initial representation vectors to obtain embedding representation vectors, selecting static neighbor regions, generating a feature map, and using a pre-trained graph convolution network to predict land usage properties, combining both static and dynamic features for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional prediction methods are used for land usage properties, then the prediction process is simple, but the prediction accuracy is insufficient
Solution Approach 1:
The patent segments the prediction task into multiple components: constructing regional correlation graphs, extracting static features from land usage data, extracting dynamic features from human interaction data, and integrating these features through a graph convolutional network. This segmentation allows each component to be optimized independently while achieving high overall prediction accuracy.
Solution Approach 2:
The patent performs preliminary actions by pre-processing land usage data and human interaction data into structured formats, pre-defining regional correlation relationships, and pre-training the graph convolutional network model before actual prediction. This preliminary preparation enables faster and more accurate predictions during deployment.
2Adaptability or versatility
If only static land usage data is used for prediction, then the data processing is simple, but the prediction cannot capture dynamic changes in urbanization
Solution Approach 1:
The patent merges static land usage data with dynamic human interaction data into a unified prediction framework. The graph convolutional network integrates both types of data, allowing the model to capture both the structural characteristics of land usage and the dynamic patterns of urbanization through human movement and interaction.
Solution Approach 2:
The patent introduces dynamics by incorporating human interaction data that changes over time into the prediction model. The graph convolutional network processes temporal sequences of human interaction data, enabling the model to adapt to changing urbanization patterns and predict future land usage properties with higher accuracy.
Data Source
AI summary
A method of determining a regional land usage property, an electronic device and a storage medium, which relate to a field of an information technology, in particular to a field of a deep learning. The method includes: acquiring a human interaction information between a plurality of regions at a specified time; updating an initial representation vector of each of the regions according to the human interaction information, so as to obtain an embedding representation vector of each of the regions; selecting a target region from the regions, and selecting a plurality of static neighbor regions within a preset range around the target region; generating a feature map of the target region according to the embedding representation vector of the target region and the embedding representation vectors of the plurality of static neighbor regions; and predicting a land usage property of the target region by using the feature map.