World Journal of Engineering Research and Technology

*Pavan Gunda, Tarini Hemanth Kumar, Muppana Sai Karthik, Mada Sai Surya
Venkata Manoj, Lella Naga Sai

ABSTRACT

The goal of developing and using a deep learning model for transferlearning-based domain- adaptive categorization is to addressperformance degradation that occurs when a model trained on onedomain is transferred to a different but related target domain. Thisstrategy lessens the need for huge labeled datasets in the target domainby employing a pretrained convolutional neural network (CNN) as afeature extractor to reuse previously learnt information. By aligningfeature representations through domain adaptation processes, thesystem improves classification performance in a variety of previouslyuntested contexts. Experiments conducted on benchmark datasets shownotable advancements in accuracy and resilience to domain alterations.

[Full Text Article] [Download Certificate] https://doi.org/10.5281/zenodo.19343412