Optimizing ML Model for Promotion Selection

A leading multinational corporation is facing the challenge of efficiently identifying suitable candidates for promotion, specifically targeting positions from managers and below across the organization. The existing procedure involves initially pinpointing potential candidates based on recommendations and past performance. The selected individuals undergo distinct training and evaluation programs tailored to the skill requirements of each department. After the program is completed, promotions are determined by factors such as training performance, key performance indicator (KPI) completion (with a threshold of over 60% completion), and other relevant criteria.

However, the current process delays the announcement of promotions until after the comprehensive evaluation, delaying a seamless transition to new roles.

In this project, we’ll create and improve the ML model’s performance in identifying the right people for promotion. We’ll improve the current promotion identification system for managerial and lower-level positions within the client’s multinational organization. We’ll utilize data science tools like pandas to load and analyze the dataset to discover the insights presented. Additionally, we’ll implement the ML algorithm using the scikit-learn library and optimization techniques using scikit-optimize to create the best ML model that can accurately identify the eligible candidates for promotion.

This optimized ML model will facilitate a quicker identification and more efficient transition for selected individuals into their new roles across different parts of the organization.