KC Two-Way Clustering Algorithms For Multi-Child Semantic Maps In Image Mining

Abstract

Image mining is now a thriving and expanding field of computer science research. Image mining is linked to the advancement of data mining in image preparation. Image mining is used to extract hidden information and in other situations where the photos do not clearly describe the situation. Image mining combines machine learning, data handling, application autonomy, and image preparation concepts. Semantic maps are used to visualize image data stored in image databases. We recommend using Multi-Child Semantic Maps to build semantic maps which fully display the image. In this study, we propose two path clustering on Multi-Child Semantic Maps (MCSM) using the K-C Means Clustering Algorithm, also known as the MCSMK-C algorithm. This algorithm causes image clustering and instructs the mining system to look at the image's top area. When mining, the MCSMK-C algorithm considers the X and Y coordinates. The system looks for groups by examining each object's territory in the database, and it saves a region if it contains more objects than the required number.