Predictive Modeling for H1B Visa Approval Using Machine Learning

Abstract

Employers in the United States can temporarily hire non-immigrant workers with the H-1B visa. People with a bachelor's degree or work experience equivalent to it are eligible for this visa, which only allows specialty workers to work in the United States. The H-1B visa is valid for three years, but it can be extended to six years. Although the H-1B visa is the most sought-after in the world, its approval rate is low. In 2019, a total of 2,000 people worldwide applied for the visa, but only 85,000 were selected, resulting in a 42% approval rate. As the US economy improves, this visa battle becomes more competitive. The employer, salary, and other factors all play a role in this decision. This strategy can be used by both the individual and the employer between applying for the visa and receiving the final decision to be informed of the outcome before it occurs because the H1-B visa category is one of the most highly sought-after ones. Taking into account all relevant factors, this project aids in predicting whether or not an individual will be granted the H1B visa. Using the Random forest algorithm, the proposed system achieved high accuracy.