World Journal of Engineering Research and Technology

Bathula Prasanna Kumar*, G. Sri Gowri, G. Sai Charitha, Ch. Usha and D. Sharon

ABSTRACT

In recent years, machine learning, particularly deep learning, has made significant moves in various domains such as computer vision, natural language processing, and speech recognition. These advancements have predominantly been driven by supervised learning techniques that rely on large amounts of labelled data. However, acquiring labelled data is often expensive, time-consuming, and sometimes impractical, especially in specialized or dynamic domains where labelling requires expert knowledge or where the data distribution changes over time. Future research directions are proposed, focusing on the integration of multimodal data, improving robustness against noisy annotations, anddeveloping scalable frameworks for real-world deployment. By advancing these methodologies, supervised learning can become more accessible and efficient, significantly broadening its applicability to resource-constrained settings.

[Full Text Article] [Download Certificate]