Hybrid job recommendation model based on professional profile using data from job boards and Machine Learning libraries

Huamán A.; Rebaza G.; Subauste D. · Proceedings of the International Multi-Conference on Society, Cybernetics and Informatics, IMSCI (2024) · DOI: 10.54808/imsci2024.01.72

Resumen

This scientific article presents that the ideal job is crucial for undergraduate software engineering students because it is related to their mental and financial health. The proposal seeks the creation of a hybrid recommendation model (collaborative filter (CF) and content-based (CBR)) of jobs with the ability to obtain the percentage of similarity between the work and the profile obtained from the student. The difficulty of software engineering students finding a job that fits their profile greatly affects these young people over time, resulting in the resulting in the loss of large amounts of offers and unemployment. The system uses machine learning methods to recommend jobs, and in turn, the percentage of similarity between the profiles is obtained. To identify these student profiles, a professional orientation test validated by a psychologist and websites with experts in technical subjects such as Testlify and TestGorilla are used. An experimental protocol was created to evaluate the effectiveness of the model in the recommendations.

Instituciones: Universidad Peruana de Ciencias Aplicadas