easy-visa-ensemble-techniques

Context:

Business communities across the United States are experiencing a high demand for skilled labor. A key challenge they face is identifying and attracting the right talent, which is crucial for maintaining competitiveness. U. S. companies seek hard-working, talented, and qualified candidates both domestically and internationally.

Under the Immigration and Nationality Act (INA), foreign workers are allowed to enter the United States for employment on a temporary or permanent basis. This act also safeguards American workers by ensuring that their wages and working conditions are not negatively affected. This is achieved by enforcing compliance with statutory requirements when U. S. employers hire foreign labor to address local workforce shortages.

The Office of Foreign Labor Certification (OFLC) manages the immigration programs, processing applications from employers who need to hire foreign workers. They grant certifications if employers can prove that there are insufficient U. S. workers available to do the job at wages that are at least equal to what is normally paid for the job in the area of intended employment.

Objective:

In fiscal year 2016, the Office of Foreign Labor Certification (OFLC) handled 775, 979 employer applications, covering 1, 699, 957 positions for both temporary and permanent labor certifications. This represented a 9% increase in processed applications compared to the previous year. With the number of applications rising annually, the review process is becoming increasingly laborious.

The growing volume of applicants each year necessitates a machine learning solution that can efficiently shortlist candidates with a higher likelihood of visa approval. Your firm, EasyVisa, has been contracted by the OFLC to provide data-driven solutions. As a data scientist, you are tasked with analyzing the provided data and using a classification model to:

• Streamline the visa approval process.
• Advise on the applicant profiles that should be approved or denied a visa, based on key factors that significantly affect the case status.

Data Description

The dataset includes various attributes related to both the employee and the employer. Below is a detailed data dictionary that explains each attribute:

• case_id : Unique ID for each visa application.
• continent : Continent where the employee is from.
• education_of_employee : Educational background of the employee.
• has_job_experience : Indicates if the employee has any job experience (Y for Yes, N for No).
• requires_job_training : Indicates if the employee needs job training (Y for Yes, N for No).
• no_of_employees : Total number of employees in the employer's company.
• yr_of_estab : Year the employer's company was established.
• region_of_employment : The intended region of employment in the US for the foreign worker.
• prevailing_wage : The average wage paid to workers in the same occupation and area, ensuring the foreign worker receives fair compensation.
• unit_of_wage : The unit for the prevailing wage, with options like Hourly, Weekly, Monthly, or Yearly.
• full_time_position : Indicates whether the job is a full-time (Y for Full Time, N for Part Time).
• case_status : Status of the visa application, whether it was certified or denied.

Here's an improved and more structured list of skills:

Skills

• Exploratory Data Analysis : Investigating datasets to discover patterns, anomalies, and test assumptions.
• Data Preparation : Preparing and cleaning data to ensure it is suitable for model building.
• Model Building :
  • Bagging : Implementing bagging techniques to reduce variance and avoid overfitting.
  • Boosting : Applying boosting methods to sequentially improve model predictions.
  • Stacking : Utilizing stacking approaches to combine predictions from multiple models.
• Hyperparameter Tuning : Optimizing model parameters to enhance performance.
• Model Performance Comparison : Evaluating and comparing the effectiveness of different models to select the best performer.

Insights

From our analysis, we can deduce that applicants who are deemed qualified tend to exhibit several distinguishing characteristics when compared to those who are not qualified:

• They typically receive a higher prevailing wage.
• They are often employed by companies that have a more substantial workforce.
• They are usually hired by companies with a well-established reputation that have been in operation for a longer period.
• They possess at least a high school level of education.
• They have previous job experience.
• They are compensated on an annual basis.

Key Conclusions

• The visa approval process should meticulously assess the prevailing wage, the number of employees, and the reputation of the hiring company, as these are critical factors in identifying qualified applicants.
• The visa approval process should generally exclude applicants who have not completed high school, due to their lower likelihood of certification, except in exceptional circumstances.

Recommendations for Business

Based on the insights gathered from our analysis, here are a set of recommendations for the visa processing bureau to refine and enhance the visa approval process:

1. Prioritize Wage Standards: Implement a more rigorous assessment of the prevailing wage offered to applicants. Those receiving wages significantly above the prevailing rate should be considered favorably, as this suggests a higher value placed on their skills and potential contribution.

2. Evaluate Employer Profile: Give additional consideration to applicants from larger and more established companies. These businesses often have more robust vetting processes and are likely to support successful integrations into the U.S. workforce.

3. Assess Company Reputation: Develop a system to evaluate the reputation and historical operation period of the sponsoring companies. Applicants from companies with a longer and positive track record should be viewed as having higher potential for contributing effectively to their fields.

4. Education Verification: Maintain a standard that all visa applicants must have at least a high school education unless there are exceptional circumstances justifying an exception. This baseline ensures a minimum qualification that could be necessary for job performance and easier integration.

5. Experience Requirement: Continue to value job experience highly in the evaluation process. Previous job experience not only shows capability but also adaptability, which is crucial for success in a new work environment.

6. Annual Compensation Focus: Favor applicants who are compensated on an annual basis. This often indicates a stable and long-term commitment by the employer, which could translate to a lower risk of visa misuse.

7. Special Case Considerations: While maintaining high standards, also develop a clear framework for handling special cases where the applicant’s unique skills or circumstances might justify an exception to the usual educational or other requirements.

Implementing these recommendations should help streamline the visa approval process, making it more efficient and aligned with the characteristics that indicate a high potential for applicant success and contribution in the United States.