Abstract

In this paper, the object identification and classification methods in a class of objects using AI based supervised and unsupervised training algorithms is presented. This paper explores object identification and classification methods within a specific class of objects using a combination of AI-based supervised and unsupervised training algorithms. The design involves a path that integrates both approaches to identify distinct patterns within the object class. In supervised learning, a labeled dataset is utilized, employing algorithms such as Support Vector Machines, Random Forests, or Convolutional Neural Networks. The model trained on this dataset learns to recognize patterns associated with different classes, enabling accurate identification of new, unlabeled data. In contrast, unsupervised learning, using algorithms like K-Means or Hierarchical Clustering, clusters objects based on inherent similarities, revealing patterns without predefined labels. The synergistic application of both methods offers a comprehensive understanding of intricate patterns within the object class, facilitating a nuanced analysis.