• workflow
• ranklib
• learning to rank
• knime analytic platform
• python
• machine learning

This thesis is about implementation and integration of learning to rank algorithms in python on KNIME Analytic Platform, resulting in the construction of a workflow capable of executing a ranking task for the human recruiting scenario. The work of thesis is developed in collaboration with the FINDHR (Fairness and Intersectional Non-Discrimination in Human Recommendation) project, and one of the requirements to meet was to provide an easy-to-use workflow for people without expertise in the programming field. The implementation of the learning to rank algorithms was achieved in two ways: with an extension of the platform that enables python code programming for complex tasks, and through the development of a custom extension, i.e., custom python-based nodes from scratch with our desired functionalities. In conclusion, the work of thesis granted a new feature on the KNIME analytic platform, that is, the ranking task, which was missing from the very start.