Using machine learning make predictions if an employee is likely to be promoted or not

Link to the competition ->>>>> https://www.kaggle.com/c/intercampusai2019/data

CASE STUDY: YAKUB TRADING GROUP - ALGORITHMIC STAFF PROMOTION Abdullah’s Baba Yakub, 38, is the heir apparent to the highly revered Yakub business dynasty. The enterprise has spanned decades with vast investment interest in all the various sectors of the economy.

Abdullah has worked for 16 years in Europe and America after his first and second degrees at Harvard University where he studied Engineering and Business Management. He is a very experienced technocrat and a global business leader who rose through the rank to become a Senior Vice President at a leading US business conglomerate. His dad is now 70 and has invited him to take over the company with a mandate to take it to the next level of growth as a sustainable legacy. Abdullah is trusted by his father and his siblings to lead this mandate.

On resumption, he had an open house with the staff to share his vision and to listen to them on how to take the business to the next level. Beyond the general operational issues and increasing need for regulatory compliance, one of the issues raised by the staff was a general concern on the process of staff promotion. Many of the staff allege that it is skewed and biased. Abdullah understood the concern and promised to address it in a most scientific way.

You have been called in by Abdullah to use your machine learning skills to study the pattern of promotion. With this insight, he can understand the important features among available features that can be used to predict promotion eligibility.

The dataset contains these variables as explained below:

• EmployeeNo : System-generated unique staff ID

• Division: Operational department where each employee works

• Qualification: Highest qualification received by the staff

• Gender: Male or Female

• ChannelofRecruitment: How the staff was recruited – this is via internal process, use of an agent or special referral

• Trainings_Attended : Unique paid and unpaid trainings attended by each staff in the previous business cycle

• Yearofbirth: Year that the employee was born

• LastPerformanceScore Previous year overall performance HR score and rated on a scale of 0-14

• Yearofrecruitment : The year that each staff was recruited into the company

• Targets_met: A measure of employees who meet the annual set target. If met, the staff scores 1 but if not, it is a 0.

• Previous_Award : An indicator of previous award won. If yes, it is a 1 and if No it is a 0.

• Trainingscoreaverage: Feedback score on training attended based on evaluation

• StateOfOrigin: The state that the employee claims

• Foreign_schooled: An indicator of staff who had any of their post-secondary education outside the country. Responses are in Yes or No

• Marital_Status: Marriage status of employees and recorded as Yes or No

• PastDisciplinaryAction : An indicator if a staff has been summoned to a disciplinary panel in the past. This is indicated as Yes or No

• PreviousIntraDepartmentalMovement : This is an indicator to identify staff who have moved between departments in the past. Yes and No are the responses.

• Noofprevious_employers : A list of the number of companies that an employee worked with before joining the organisation. This is recorded as counts