Use of machine learning techniques in human resource management
Loading...
Date
2026-05-22
Authors
Συμεωνίδης, Ανδρέας
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis presents the design, development, and evaluation of iHR (Intelligent HR), a specialised software platform aimed at modernising Human Resource Management through the rigorous analysis of business data. Aiming to tackle the limitations of traditional, intuition-based management in the HR 4.0 era, this project delivers an integrated solution that centralises and analyses data within its Learning and Development (L\&D) module. The application incorporates a hybrid recommendation engine, automated sentiment analysis, and interactive data visualisations to bridge the gap between friction and insight, identified in existing market solutions.
A requirement-gathering approach, involving structured consultations with HR professionals at an international corporation, guided the development process. This ensured the platform met critical organisational needs for talent profiling, automated feedback processing, and standardised skill-gap identification. The system architecture leverages a decoupled N-tier model, utilising FastAPI for the backend, Next.js for the reactive frontend, and MySQL for relational data persistence. Furthermore, the platform adheres to Privacy by Design principles, ensuring GDPR compliance while handling sensitive information.
Evaluation of the system confirms that iHR effectively eliminates data silos and provides an objective, data-driven foundation for employee calibration. Through the integration of the ESCO dataset and VADER sentiment analysis, the platform transforms qualitative feedback into measurable metrics, enhancing corporate training quality and transparency. Future developments can include the vertical expansion of L\&D modules, the integration of performance and wellness tracking, and the transition to a multi-tenant SaaS model to support Small-to-Medium Enterprises (SMEs).
Description
Keywords
Human-computer interaction, Full stack development, Machine learning, Natural language processing, Python, FastAPI, Next.JS