Data Driven Employee Promotion Facilitation
Background
HR analytics is revolutionising the way human resources departments operate, leading to higher efficiency and better results overall. Human resources has been using analytics for years. However, the collection, processing and analysis of data has been largely manual, and given the nature of human resources dynamics and HR KPIs, the approach has been constraining HR.
One challenging HR related problem a multinational company facing with is to identify the right people for promotion for mid-level manager position and prepare them in time.
Identify the eligible candidates at a particular checkpoint during the evaluation program so that the entire promotion cycle could be speed up. The task is to predict whether a potential promotee at checkpoint (at about 60% of the process) will be promoted or not after the evaluation process.
Problem Statement
Current (less efficient) process steps:
-
1
Identify a set of employees based on recommendations or past performance.
-
2
Selected employees go through the separate training and evaluation program based on the required skills.
-
3
At the end of the program, based on various factors (training performance, KPI completion etc.) successful employees get promotion.
-
4
For the abovementioned process, the final promotions are only announced after the evaluation and this leads to delay in transition to their new roles.
Data Description
Data Set Size
50k+ labeled points
Descriptors
Attributes around employees' past and current performance along with demographics
Target
If employee will be promoted (binary classification)
Solution
Data Preparation
Missing Data Imputation, Feature Encoding, Feature Combination
Modeling
Boosted Tree Classifier (CatBoost) with 10-fold Cross Validation
Performance
Accuracy (84%)
F1-Score(0.42)
Precision (0.3)
Recall (0.69)
Deployment
Containerized microservice (REST API)