Browsing by Author "Novais, Liliana Cristina de Lemos"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
- CurEval - Curriculum EvaluationPublication . Novais, Liliana Cristina de Lemos; Martins, António Constantino LopesEfficiently screening and evaluating curricula in recruitment processes is a critical task that often requires substantial time and effort from Human Resources professionals. This work presents CurEval, an algorithm developed to automate the evaluation and screening of curricula based on vacancy requirements. The algorithm utilizes a predefined set of keywords and a CSV file format for input, facilitating easy data structuring and processing. To validate the algorithm’s performance and address privacy concerns, synthetic curricula were generated using templates with slight variations in personal data. The algorithm’s results were compared with evaluations made by a Human Resources collaborator and external paid recruitment platforms. The study’s findings indicate that CurEval effectively filters out irrelevant curricula, reducing the screening workload for HR professionals. The algorithm aligns with human evaluations, ensuring accurate classification of curricula according to vacancy requirements. Additionally, bias analysis revealed no evidence of discriminatory bias in the algorithm or human evaluations in the sample data. Further improvements for CurEval include expanding the list of keywords, incorporating natural language processing techniques, and integrating machine learning to enhance accuracy and adaptability. Real-time data integration, feedback loops with HR professionals, and integration with Applicant Tracking Systems are suggested to streamline the recruitment process. Multi-lingual support, performance metrics, and ongoing ethical considerations are also essential for refining and maintaining the algorithm’s effectiveness and fairness. CurEval offers promising potential to revolutionize the curricula evaluation process, enabling faster and more efficient screening while ensuring fairness and equal opportunity. Future work should focus on enhancing the algorithm’s capabilities, addressing biases, and continuously validating and improving its performance through collaboration and feedback from HR professionals.