Integration of Land and Tax Geospatial Data with Artificial Intelligence (AI)-Based Clustering

Integrating Land Geospatial Data and Tax Geospa-tial Data is challenging due to differences in file formats and attributes. This research addresses these challenges and seeks to effectively combine the datasets. Initially, the study identifies a suitable file format that can handle the various attributes of both datasets without losing data quality. A systematic approach is then employed to align the different attributes. The research focuses on data from East Jakarta, Indonesia. The land data includes files such as. shp,. shx,. dbf, and. prj, representing the shapes and attributes of land parcels. The tax data includes sim-ilar files containing detailed tax information. Since the datasets lack common attributes, spatial relationships are used to join them using QGIS software. The process begins by simplifying the data structures and clipping them to the relevant areas. Spatial joins are performed based on where the data overlaps the most. Post-integration, the dataset is cleaned to remove any duplicate or unnecessary information. K-Means clustering with Scikit-Iearn is then used to analyze the data and identify patterns. The results show that land and tax data can be combined spatially, but not by their attributes. When joined, the tax data, which may be divided among multiple owners, follows the land parcel data, resulting in a single ownership representation. This method ensures that the datasets are effectively merged, enabling better analysis and decision-making. The approach demonstrates a robust methodology for integrating disparate geospatial datasets, enhancing data accessibility and utility.