Classification and Evaluation of Indian Faces Using Machine Learning Models
Madala Veerendra Kumar, Pullela Harish Chowdary, Chetna Vaid Kwatra, Talari Nandeesh, Kommu Venkata Pawan · 2024
In a country like India whose citizens carry a diversity of facial features and appearances widely varying from each other this paper can address one of the solutions that can be considered accountable to classify and distinguish the differences among each other. Facial classification is crucial for computer vision applications and Machine learning programs such as image retrieval, audit systems, biometrics, and security mechanisms. The purpose of this paper is to express our findings and understanding of Indian face classification based on states using Python. Our approach to the idea is a Machine Learning model that is trained on a compact dataset of Indian faces labeled with their respective states. The model has achieved results of 42%,40%,42%, and 30% respectively on benchmarks of accuracy, precision, recall, and f1 score. The approach to this project has key phases of data collection and a Novel dataset creation, cumulating data, extracting features, and feeding the results as input to SVM and Logistic regression models to classify the data. This work contributes to further research and development of innovative applications that can benefit the field of research.